Information processing device, information processing method, and information processing program

Abstract

Provided is an information processing device comprising: a data generation unit that generates generation data including generation explanatory variables and generation objective variables on the basis of reference data indicating relationships between reference explanatory variables and reference objective variables corresponding to the reference explanatory variables; and a data output unit that, on the basis of the generation objective variables and a predetermined output condition, generates output data including the generation explanatory variables, and outputs the generated output data. The data output unit may optionally output the generation objective variables to a plurality of users of the generation data, and output the generation explanatory variables to at least one user on the basis of the output condition and user conditions presented by the respective users in order to acquire the generation explanatory variables corresponding to the output generation objective variables.

Description

Information processing device, information processing method, and information processing program

[0001] The present invention relates to an information processing device, an information processing method, and an information processing program.

[0002] Patent Document 1 describes "plurality of models generated for each of a plurality of financial institutions based on financial transaction information of customers held by each of the financial institutions" (abstract). Patent Document 2 describes "receiving parameters related to the operating status of semiconductor manufacturing equipment in an anonymized format" (abstract). Patent Document 3 describes "registering anonymized data obtained by anonymizing customer information acquired from a customer information management device in a first database" (abstract). [Prior Art Documents] [Patent Documents] [Patent Document 1] WO 2023 / 281685 [Patent Document 2] WO 2023 / 053161 [Patent Document 3] WO 2022 / 137561 General disclosure

[0003] One embodiment of the present invention relates to a system intended for use in materials informatics using machine learning models. When a machine learning model (e.g., secure computing, federated learning) generated by combining data from multiple data providers is shared among multiple users, it is conceivable that each user will input similar values ​​as target values. In such a case, if reverse analysis of the model is performed based on the target values ​​input by each user, the computer will repeatedly perform similar processing, resulting in ineffective use of computing resources.

[0004] In one embodiment of the present invention, a "data generation unit" generates "optimization data," which are explanatory variables corresponding to the target values ​​of the objective variables, and a "data output unit" determines to which user the data should be output. At this time, the determination is made using "user conditions," which are conditions related to the target values ​​of materials informatics and the like, acquired in advance from the user. With this configuration, even when similar target values ​​are input from multiple users, it is possible to output optimization data to each user with a small number of calculations, and by preventing duplicate calculations, it is possible to make effective use of computational resources.

[0005] A first aspect of the present invention provides an information processing device comprising: a data generation unit that generates generation data including generation explanatory variables and generation objective variables based on reference data indicating relationships between reference explanatory variables and reference objective variables corresponding to the reference explanatory variables, and a data output unit that generates output data including the generation explanatory variables based on the generation objective variables and predetermined output conditions, and outputs the generated output data.

[0006] The data output unit may output the generation objective variables to multiple users of the generation data, and output the generation explanatory variables to at least one user based on the user conditions and output conditions presented by each user in order to obtain the generation explanatory variables corresponding to the output generation objective variables.

[0007] In any of the above information processing devices, when the data output unit outputs a generation explanatory variable to at least one user whose user conditions presented by the user in order to obtain the generation explanatory variable satisfy the output conditions, the data output unit does not need to output a generation objective variable corresponding to the output generation explanatory variable to other users for a predetermined period of time.

[0008] In any of the above information processing devices, the data generation unit may generate generation data including a plurality of generation explanatory variables and a plurality of generation target variables corresponding to the plurality of generation explanatory variables, respectively. The data output unit may not output the output generation explanatory variables and other generation explanatory variables within a predetermined similarity range to the output generation explanatory variables to other users for a predetermined period of time.

[0009] In any of the above information processing devices, the data generation section may determine the similarity range based on the degree of similarity between one generation explanatory variable and another generation explanatory variable.

[0010] In any of the above information processing devices, the data generating section may determine the similarity range further based on the degree of similarity between one generation target variable and another generation target variable.

[0011] Any of the above information processing devices may further include a storage unit that stores user conditions presented by each of a plurality of users of the generation data in order to acquire the generation objective variables. The data output unit may output the generation explanatory variables to at least one user when the user conditions of at least one user satisfy the output conditions for the generation objective variables that at least one user wishes to acquire.

[0012] In any of the above information processing devices, the user conditions may include an amount to be paid by the user to acquire the generation objective variable and the generation explanatory variable. When the user conditions of each of the multiple users satisfy the output conditions, the data output unit may determine one user to output the generation explanatory variable based on the amount.

[0013] In any of the above information processing devices, the data output section may conceal the user conditions of one user from other users.

[0014] In any of the information processing devices described above, the data output unit may output, to one user of the generated data, information indicating other generation objective variables that are recommended for acquisition, based on a generation objective variable corresponding to a generation explanatory variable output to the one user.

[0015] In any of the above information processing devices, one exclusive range for the generation object variable may be determined in advance for one user of the generated data. Another exclusive range for the generation object variable may be determined in advance for another user of the generated data. If a generation object variable that one user wishes to acquire belongs to another exclusive range, the data output unit does not need to output the generation object variable that belongs to the other exclusive range to the one user.

[0016] In any of the above information processing devices, one exclusive range for the generation object variable may be predetermined for one user of the generated data. Another exclusive range for the generation object variable may be predetermined for another user of the generated data. If the generation object variable that one user wishes to acquire belongs to another exclusive range, the data output unit may determine that the generation object variable belonging to the other exclusive range does not satisfy the output condition.

[0017] In any of the above information processing devices, the reference explanatory variables and the reference objective variable may be provided by a data provider. When the data output unit outputs the generation explanatory variables to a user of the generation data, the data output unit may output a reward to the data provider who provided the reference explanatory variables based on a distance between the output generation explanatory variables and the reference explanatory variables.

[0018] In any of the above information processing devices, the data generation unit may have a first inference model that outputs a first predicted value of a generation objective variable corresponding to an input explanatory variable when an input explanatory variable is input by machine learning first reference data having a reference explanatory variable and a reference objective variable provided by each of a plurality of data providers. When a target value of the generation objective variable is input to the data generation unit, the data generation unit may output optimization data that is a generation explanatory variable corresponding to the target value.

[0019] In any of the information processing devices described above, the data generation unit may generate candidates for optimization data in response to an input of a target value. The first inference model may output a first predicted value corresponding to the generated candidates. The data generation unit may output one of the candidates as optimization data based on the first predicted value and the target value.

[0020] In any of the above information processing devices, the data generation unit may have a second inference model that outputs a second predicted value of the generation objective variable when the input explanatory variables are input, by machine learning second reference data obtained by excluding, from the first reference data, individual explanatory variables that are reference explanatory variables corresponding to at least one of the multiple data providers and individual objective variables that are reference objective variables corresponding to at least one of the multiple data providers. Any of the above information processing devices may further include a contribution calculation unit that calculates a contribution of the first reference data corresponding to at least one data provider to the generation explanatory variables based on the first predicted value and the second predicted value.

[0021] In any of the above information processing devices, the reference explanatory variables may include a plurality of types of parameters. At least one type of parameter may be provided by at least one data provider. Any of the above information processing devices may further include a contribution calculation unit that calculates a contribution of the at least one parameter to the optimization data based on the optimization data and the first reference data.

[0022] In any of the above information processing devices, the data generation unit may perform machine learning on second reference data for each of the multiple data providers, from which individual explanatory variables and individual objective variables corresponding to one of the multiple data providers have been excluded. The second inference model may output a second predicted value for each of the multiple data providers. The contribution calculation unit may calculate a contribution for each of the multiple data providers based on the first predicted value and the second predicted values ​​for each of the multiple data providers. When the data output unit outputs the generation explanatory variables to a user of the generation data, it may output a reward to each of the multiple data providers in accordance with the contribution of each of the multiple data providers to the output generation explanatory variables.

[0023] In any of the information processing devices described above, the data generation unit may include a third inference model that outputs a third predicted value of the generated objective variable when an input explanatory variable is input by machine learning third reference data obtained by excluding one reference explanatory variable and one reference objective variable corresponding to the one reference explanatory variable from the first reference data. Any of the information processing devices described above may evaluate the reliability of the first reference data based on the first predicted value and the third predicted value.

[0024] In any of the above information processing devices, the data output section may determine, based on the reliability, whether or not to output the generation target variable generated by the data generation section to a user of the generated data.

[0025] Any of the information processing devices described above may further include an input unit. When a user of the generated data inputs to the input unit a request that the data output unit output new generation explanatory variables, the data generation unit may generate new generation data.

[0026] Any of the above information processing devices may further include an input unit. The reference explanatory variables and the reference objective variable may be provided by each of a plurality of data providers. When an individual explanatory variable that is a reference explanatory variable corresponding to at least one of the plurality of data providers and an individual objective variable that is a reference objective variable corresponding to at least one of the data providers are newly input to the input unit, the data generation unit may generate new generation data.

[0027] A second aspect of the present invention provides an information processing method, comprising: a data generating step in which a data generating unit generates generated data including generated explanatory variables and generated objective variables based on reference data indicating relationships between reference explanatory variables and reference objective variables corresponding to the reference explanatory variables; and a data output step in which a data output unit generates output data including the generated explanatory variables based on the generated objective variables and predetermined output conditions, and outputs the generated output data.

[0028] In a third aspect of the present invention, there is provided an information processing program that causes a computer to function as an information processing device.

[0029] The above summary of the invention does not list all of the features of the present invention, and subcombinations of these features may also be inventions.

[0030] 1 is a diagram showing an example of an information providing system 500. FIG. 1 is a diagram showing an example of reference data 201 provided by a data provider 200. FIG. 2 is a diagram showing an example of first reference data 210 provided by each of a plurality of data providers 200. FIG. 3 is a diagram showing an example of an information processing device 100 according to an embodiment of the present invention. FIG. 4 is a diagram showing an example of a data space 212. FIG. 5 is another diagram showing an example of a data space 212. FIG. 6 is a diagram showing an example of learning by a data generating unit 20. FIG. 7 is a diagram showing another example of learning by a data generating unit 20. FIG. 8 is a diagram showing another example of inference by a second inference model 24. FIG. 9 is a diagram showing another example of inference by a second inference model 24. FIG. 10 is a diagram showing another example of learning by a data generating unit 20. FIG. 11 is a diagram showing another example of first reference data 210 provided by each of a plurality of data providers 200. A flowchart showing an example of an information processing method according to an embodiment of the present invention. A flowchart showing an example of an information processing method according to an embodiment of the present invention. A flowchart showing an example of an information processing method according to an embodiment of the present invention. A flowchart showing an example of an information processing method according to an embodiment of the present invention. A flowchart showing an example of an information processing method according to an embodiment of the present invention. 2 is a flowchart showing another example of an information processing method according to an embodiment of the present invention. FIG. 3 is a diagram showing another example of the configuration of the information processing device 100. FIG. 4 shows an example of information on the contribution of each explanatory variable calculated by the contribution calculation section 60. FIG. 5 shows another example of information on the contribution of each explanatory variable calculated by the contribution calculation section 60. FIG. 6 is a diagram explaining an example of the operation of the plan formulation section 72. FIG. 7 is a diagram explaining another example of the operation of the plan formulation section 72. FIG. 8 is a diagram showing another example of the configuration of the information processing device 100. FIG. 9 is a diagram showing an example of a computer 2200 in which the information processing device 100 according to an embodiment of the present invention may be embodied in whole or in part.

[0031] The present invention will be described below through embodiments of the invention, but the following embodiments do not limit the scope of the invention as claimed. Furthermore, not all of the combinations of features described in the embodiments are necessarily essential to the solution of the invention.

[0032] 1 is a diagram showing an example of an information providing system 500. The information providing system 500 includes an information processing device 100, a network 400, a plurality of terminals 300, and a plurality of terminals 360. The information processing device 100 in this example is a server that provides services provided by the information providing system 500. The terminal 300 in this example is a terminal of a data provider 200 that receives services provided by the information providing system 500. The terminal 360 in this example is a terminal of a user 260 that receives services provided by the information providing system 500.

[0033] In this example, data providers 200-1 to 200-n (n is an integer of 2 or greater) have terminals 300-1 to 300-n, respectively, and users 260-1 to 260-n' (n' is an integer of 2 or greater) have terminals 360-1 to 360-n', respectively. The numbers n and n' may be different or the same. Network 400 may be a general-purpose network such as the Internet, or a network using dedicated lines.

[0034] The data provider 200 and the user 260 may be different or may be the same. In this example, the data provider 200 and the user 260 are different.

[0035] The terminal 300 transmits the individual reference data 202 provided by the data providers 200 to the information processing device 100 via the network 400. The information processing device 100 receives the individual reference data 202 transmitted from each of the multiple terminals 300 and stores the received individual reference data 202. One individual reference data 202 provided by one data provider 200 and stored in the information processing device 100 may be kept secret from other data providers 200.

[0036] The information processing device 100 may generate the estimation model 21 using the individual reference data 202 transmitted from each of the multiple terminals 300. The estimation model 21 may be a regression model such as an arithmetic expression generated by regression analysis or the like, or may be a model generated by machine learning.

[0037] The individual reference data 202 has individual explanatory variables 203 (described later) and individual objective variables 205 (described later) corresponding to the individual explanatory variables 203. The individual reference data 202 indicates the relationship between the individual explanatory variables 203 and the individual objective variables 205. The individual explanatory variables 203 may be reference explanatory variables 204 (described later) corresponding to at least one of the multiple data providers 200. The individual objective variables 205 may be reference objective variables 206 (described later) corresponding to at least one of the multiple data providers 200.

[0038] The individual explanatory variables 203 (described later) and the reference explanatory variables 204 (described later) are explanatory variables that indicate, for example, the state of a material or the like before some processing is performed. The individual objective variable 205 (described later) and the reference objective variable 206 (described later) are objective variables that indicate, for example, the state of a product or the like after some processing is performed. The estimation model 21 is a model that indicates the relationship between the reference explanatory variables 204 and the reference objective variable 206. The estimation model 21 is a model that can estimate a corresponding objective variable when an explanatory variable is input.

[0039] The terminal 360 may transmit user conditions 262 desired by the user 260 to the information processing device 100 via the network 400. The information processing device 100 may receive the user conditions 262 transmitted from each of the multiple terminals 360. The user conditions 262 are conditions presented by each user 260 in order to acquire a generation explanatory variable 224 (described below). The user conditions 262 include at least one of the following: the type of generation objective variable 226 (described below) that the user 260 wishes to acquire, the value of the generation objective variable 226, the value range of the generation objective variable 226, the desired monopoly period for the generation objective variable 226, and a reward to be paid to at least one of the multiple data providers 200 for acquiring the generation objective variable 226. The reward is an index representing the value of the generation objective variable 226 acquired by the user 260. The index may be, for example, money. If the user 260 desires to acquire multiple generation objective variables 226, the user condition 262 may include a priority among the multiple generation objective variables 226. The user condition 262 may be different for each of the multiple users 260.

[0040] The information providing system 500 may further include an intermediation unit 420. The intermediation unit 420 is, for example, an agent in which the user conditions 262 are registered. The intermediation unit 420 may be an API (Application Programming Interface). When the intermediation unit 420 is an API, the intermediation unit 420 may be provided by the user 260. The users 260-1 to 260-n' may provide the intermediation units 420-1 to 420-n', respectively. The user 260 may register the user conditions 262 in the intermediation unit 420 via the terminal 360.

[0041] The information processing device 100 of this example can generate a highly accurate estimation model 21 by generating the estimation model 21 using individual reference data 202 from multiple data providers 200. Each data provider 200 can receive services based on the highly accurate estimation model 21 without disclosing the individual reference data 202 that it has provided to the other data providers 200.

[0042] FIG. 2 is a diagram showing an example of reference data 201 provided by a data provider 200. The reference data 201 has reference explanatory variables 204 and reference objective variables 206. In this example, the reference explanatory variables 204 include one or more physical quantities. In this example, the reference objective variables 206 are physical quantities realized using the physical quantities indicated by the reference explanatory variables 204. For example, the reference objective variables 206 are physical property values ​​of an object to be manufactured. In this example, the reference explanatory variables 204 include at least one of a parameter indicating the material of the object to be manufactured and a parameter indicating the processing conditions in a manufacturing process using the material. The reference explanatory variables 204 shown in FIG. 2 include parameters related to a reaction process R and a molding process M. In this example, the parameters related to the reaction process R include a composition Co and a process condition P1.

[0043] The composition Co may include the types and composition ratios of one or more raw materials. In the example of FIG. 2, the composition Co includes the types and composition ratios of the raw materials of the main material Sm, the sub-material Sb, and the additive Sa. The numerical values ​​shown in FIG. 2 are an example of the composition ratios. The process condition P1 may include reaction conditions for the material having the composition Co. In the example of FIG. 2, the process condition P1 includes a process temperature and a process time in the reaction treatment of the material having the composition Co. In this example, the molding process M includes a process condition P2. The process condition P2 may include a process temperature and a process pressure in the process of molding the substance produced in the reaction process R into a predetermined shape.

[0044] The reference explanatory variable 204 may have evaluation conditions Ev in the evaluation process E. The evaluation process E in this example is a process for evaluating the physical property values ​​of the molded product molded in the molding process M. The evaluation conditions Ev are conditions for evaluating the molded product molded in the molding process M in the evaluation process E. The evaluation conditions Ev may include an evaluation temperature and an evaluation humidity, which are the temperature and humidity of the measurement environment in which the physical property values ​​of the molded product are measured.

[0045] The reference objective variable 206 is a parameter related to the evaluation result when the molded product molded in the molding process M is evaluated under the evaluation condition Ev. In this example, the reference objective variable 206 includes the evaluation results of the hardness and elongation of the molded product molded in the molding process M. The reference data 201 in this example is actual data obtained when the data provider 200 actually produces a molded product. At least a part of the reference data 201 may be data obtained by calculation such as a simulation.

[0046] 3 is a diagram showing an example of first reference data 210 provided by each of multiple data providers 200. The first reference data 210 is an example of reference data 201 (see FIG. 2). The multiple data providers 200 are, for example, multiple companies belonging to the same industry. The individual reference data 202 is the reference data 201 (see FIG. 2) provided by one data provider 200. In this example, individual reference data 202-1 to individual reference data 202-n are provided by data providers 200-1 to 200-n, respectively.

[0047] The individual reference data 202 has individual explanatory variables 203 and an individual objective variable 205. The individual explanatory variables 203 are reference explanatory variables 204 corresponding to at least one of the multiple data providers 200. The individual objective variable 205 is a reference objective variable 206 corresponding to at least one of the multiple data providers 200. The individual reference data 202-u includes an individual explanatory variable 203-u and an individual objective variable 205-u (where u is an integer from 1 to n). The first reference data 210 is data obtained by aggregating the individual reference data 202-1 to 202-n. The individual reference data 202-1 to 202-n may have the same parameters.

[0048] The first reference data 210 may include an identification parameter Id for identifying the multiple data providers 200. In this example, the identification parameter Id includes the name and data number of the data provider. In this example, the individual reference data 202-1 to 202-n provided by the data providers 200-1 to 200-n are labeled "Company A" to "Company N," respectively.

[0049] One data number may correspond to one set of the reference explanatory variables 204 and the reference objective variables 206. In this example, the three sets included in the individual reference data 202-1 are assigned data numbers "A1" to "A3", the three sets included in the individual reference data 202-2 are assigned data numbers "B1" to "B3", and the three sets included in the individual reference data 202-n are assigned data numbers "N1" to "N3".

[0050] 4 is a diagram showing an example of an information processing device 100 according to an embodiment of the present invention. The information processing device 100 includes a data generation unit 20 and a data output unit 40. The information processing device 100 may include an identification unit 10, an input unit 30, a storage unit 50, a contribution calculation unit 60, and a reliability evaluation unit 70.

[0051] A part or the whole of the information processing device 100 may be realized by a computer. The data generation unit 20 may be a CPU (Central Processing Unit) of the computer. When the information processing device 100 is realized by a computer, a program for causing the computer to function as the information processing device 100 may be installed on the computer, and a program for causing the computer to execute the information processing method described below may be installed on the computer.

[0052] The input unit 30 receives input of individual reference data 202-1 to 202-n. The input unit 30 may communicate with the terminals 300 of the data providers 200 via a network 400. The input unit 30 may receive the individual reference data 202 from the terminals 300 of the respective data providers 200. The input unit 30 may read the individual reference data 202 from a storage unit 50 that stores the individual reference data 202. The input unit 30 may have a mouse, a keyboard, or the like for inputting the individual reference data 202. The storage unit 50 may store at least one of the individual reference data 202 and the first reference data 210.

[0053] User conditions 262-1 to 262-n' are input to the input unit 30. The input unit 30 may communicate with the terminals 360 of the users 260 via the network 400. The input unit 30 may receive the user conditions 262 from the terminals 360 of the respective users 260. The user conditions 262 may be stored in the storage unit 50. The input unit 30 may read out the user conditions 262 from the storage unit 50 in which the user conditions 262 are stored.

[0054] The data generation unit 20 generates generation data 220 including generation explanatory variables 224 and generation objective variables 226 based on reference data 201 indicating the relationship between reference explanatory variables 204 and reference objective variables 206. The data generation unit 20 may generate generation data 220 including a plurality of generation explanatory variables 224 and a plurality of generation objective variables 226 corresponding to each of the plurality of generation explanatory variables 224, based on the reference data 201. For example, based on a target value Vt of the generation objective variable 226 input from a terminal 300 of any data provider 200 or input from a terminal 360 of any user 260, the data generation unit 20 generates optimization data Dt, which are generation explanatory variables 224 corresponding to the target value Vt. For example, the target value Vt of the generation objective variable 226 is a physical property value that a newly generated objective object should have. The data generating unit 20 may use the individual reference data 202 stored in the storage unit 50 to estimate a combination of explanatory variable values ​​that can achieve the target value Vt of the generation objective variable 226. The data generating unit 20 may generate optimized data Dt by inputting the target value Vt of the generation objective variable 226 into the estimation model 21. The data generating unit 20 may generate the optimized data Dt based on multiple user conditions 262. For example, the data generating unit 20 may generate the target value Vt by identifying similar user conditions 262 from at least some of the multiple user conditions 262. The data generating unit 20 may determine the similarity between the user conditions 262 using an index such as cosine similarity when the user conditions 262 are expressed as vectors. The data generating unit 20 may generate the optimized data Dt in response to the user conditions 262 satisfying predetermined conditions. For example, the data generator 20 may generate the optimized data Dt when the number of inputs of the user conditions 262 (e.g., the number of user conditions 262-1 to 262-n' input to the input unit 30 within a preset period) exceeds a threshold. Furthermore, as described above, the data generator 20 may generate the optimized data Dt when the number of inputs of similar user conditions 262 exceeds a threshold.

[0055] The identification unit 10 identifies a data space 212 (described later) of each of the multiple data providers 200 based on the individual reference data 202 provided by each of the multiple data providers 200. The data space 212 (described later) for each data provider 200 is a portion of a data space 214 (described later) of all participants. The data space 212 is a virtual space. As an example, the data space 212 (described later) is a multidimensional space having axes corresponding to each parameter included in the individual reference data 202. Each individual reference data 202 is placed at a position in the data space 214 (described later) of all participants depending on the value of each parameter. The data space 212 (described later) of each data provider 200 may include all of the individual reference data 202 provided by the data provider 200. The data space 212 (described later) of each data provider 200 may also include the area between any two individual reference data 202 among the multiple individual reference data 202 provided by the data provider 200. In addition, the identification unit 10 may refer to the individual reference data 202 provided by each of the multiple data providers 200 and identify which data provider 200 provided each individual reference data 202.

[0056] The data generation unit 20 may generate optimized data Dt corresponding to the target value Vt based on the target value Vt of the generation target variable 226 and data space information Is, which is information indicating the data space. For example, the data generation unit 20 generates optimized data Dt based on the data space information Is of a data provider 200 that inputs the target value Vt of the generation target variable 226. The data generation unit 20 may generate optimized data Dt that is not included in the data space 212 (described below) of the data provider 200.

[0057] FIG. 5 is a diagram showing an example of the data space 212. For ease of explanation, the case where n in FIG. 3 is 2 will be described. The identification unit 10 may identify the data space 212 of each of the multiple data providers 200 based on the individual reference data 202. In this example, the identification unit 10 identifies the data space 212-1 of the data provider 200-1 (Company A in FIG. 3) based on the individual reference data 202-1, and identifies the data space 212-2 of the data provider 200-2 (Company B in FIG. 3) based on the individual reference data 202-2. In FIG. 5, the data space 212-1 is indicated by a fine dashed line, and the data space 212-2 is indicated by a coarse dashed line. The identification unit 10 may identify the data space 212 based on the individual explanatory variables 203, or may identify the data space 212 based on the individual explanatory variables 203 and the individual objective variable 205.

[0058] The identification unit 10 may identify a data space 214 based on all data of the individual reference data 202-1 to 202-n. For example, the data space 214 of all participants is a space having axes corresponding to all parameters appearing in at least one individual reference data 202. When there are two individual reference data 202 (individual reference data 202-1 and individual reference data 202-2), the data space 214 is the outer frame (thick frame in FIG. 5) of the union of the data space 212-1 (fine dashed frame in FIG. 5) and the data space 212-2 (coarse dashed frame in FIG. 5). Within the data space 212, there are data spaces 212-1 to 212-n of each data provider 200. FIG. 5 shows an example where n is 2.

[0059] Data space 212-1 is designated area D1, and data space 212-2 is designated area D2. As shown in FIG. 5, there may be an overlapping area between data space 212-1 and data space 212-2. This overlapping area is designated overlapping area D3. In FIG. 5, overlapping area D3 is indicated by hatching. The area of ​​area D1 excluding overlapping area D3 is an area not covered by data space 212-2 of data provider 200-2 (Company B in FIG. 3). The area of ​​area D2 excluding overlapping area D3 is an area not covered by data space 212-1 of data provider 200-1 (Company A in FIG. 3).

[0060] FIG. 6 is another diagram showing an example of the data space 212. As with FIG. 5, the case where n is 2 will be described. For ease of illustration, FIG. 6 shows the data space 212 when there are two types of reference explanatory variables 204 (the composition ratio of raw material A in FIG. 3 and the process temperature under process condition P1). In this case, the data space 212 has an axis indicating the composition ratio of raw material A and an axis indicating the process temperature. Each of the three white circles in FIG. 6 corresponds to data numbers A1 to A3 in FIG. 3, respectively, and each of the three black circles corresponds to data numbers B1 to B3 in FIG. 3, respectively. Each data space 212 contains more individual reference data 202, but these are omitted in FIG. 6. In the example of FIG. 6, the data space 212 is a planar region based on two types of reference explanatory variables 204. In FIG. 6, data space 212-1 is indicated by a fine dashed line, and data space 212-2 is indicated by a coarse dashed line.

[0061] Each data space 212 includes all the individual reference data 202 provided by the corresponding data provider 200. The data space 212 may also include a region enclosed by connecting with straight lines any plurality of the individual reference data 202 provided by the corresponding data provider 200. The data space 212 may also be a region enclosed by connecting with straight lines the outermost plurality of the individual reference data 202 provided by the corresponding data provider 200. The data space 212 may also be a region estimated based on the individual reference data 202 by a machine learning model such as the k-nearest neighbor method or One-Class Support Vector Machine (OCSVM) used for a region estimation problem. For example, in the k-nearest neighbor method, the data generation unit 20 calculates the distance d as the average value of k distances d' between an arbitrary data point and each of a plurality of individual explanatory variables 203 in the data space 212. The data generation unit 20 may select the k distances d' in ascending order of distance. The data generation unit 20 may specify, as the data space 212, a region in which the distance d is within an arbitrary threshold. The distance d' may be the Euclidean distance between an arbitrary data point and each of the individual explanatory variables 203 in the data space 212, or may be the Mahalanobis distance.

[0062] The data generation unit 20 may generate optimized data Dt based on the target value Vt of the generation objective variable 226 and the data space information Is. The target value Vt of the generation objective variable 226 may be a desired target value of the user 260 or a desired target value of the data provider 200. In the examples of FIGS. 2 and 3 , the target value Vt of the generation objective variable 226 is a physical property value of the target object that one user 260 desires to obtain, and is at least one of the values ​​of “hardness” and “stretchability” in the evaluation step E. The target value Vt of the generation objective variable 226 may be different for each user 260. The user 260 may input the target value Vt into the input unit 30 via the terminal 360. The target value Vt may be included in the user conditions 262. As described below, the generated optimized data Dt may be output by the data output unit 40 to at least one user 260 who has presented user conditions 262 that satisfy the output conditions. The optimization data Dt may be displayed on the display unit of the terminal 360 of the at least one user 260 .

[0063] The data space information Is is information related to the data space 212. The data space information Is includes information such as the range occupied by each data space 212 in the data space 214, the size of the data space 212, the density of the reference explanatory variables 204 in the data space 212, the machine learning model for estimating the data space 212, and the maximum, minimum, and average values ​​of the distances between each data space 212. The optimized data Dt is an explanatory variable corresponding to the target value Vt. A candidate for the optimized data Dt is referred to as a candidate Ve. Multiple candidates Ve may be generated for one target value Vt.

[0064] For example, the data generator 20 generates the optimized data Dt within a range that is not included in the specific data space 212. The data generator 20 may generate the optimized data Dt within a range in which the distance from the specific data space 212 is equal to or greater than a predetermined value. The distance between the data space 212 and the specific explanatory variable will be described later. The specific data space 212 is, for example, the data space 212 of the user 260 who inputs the target value Vt. The data generator 20 may generate the optimized data Dt from an overlapping region D3 (see FIG. 5 ) within the specific data space 212.

[0065] The data generating unit 20 may generate one or more candidates Ve for the optimization data Dt based on the first reference data 210 (see FIG. 3 ). The data generating unit 20 may generate the candidates Ve based on the first reference data 210 stored in the storage unit 50. The candidates Ve are candidates for the optimization data Dt, which may be explanatory variables corresponding to the target value Vt that a user 260 desires to obtain. The data generating unit 20 may generate the candidates Ve based on the first reference data 210 and the target value Vt. The data generating unit 20 may set as the candidates Ve one or more explanatory variables obtained by inputting the target value Vt into the estimation model 21 generated from the first reference data 210. The candidates Ve may include all parameters in the reference explanatory variables 204 (see FIGS. 2 and 3 ).

[0066] For example, when the reference explanatory variable 204 has p types of parameters (for example, 15 types from "raw material A" to "temperature" of the evaluation condition Ev in FIG. 3) and the reference objective variable 206 has one type of parameter (for example, "hardness" in FIG. 3), the p types of parameters of the reference explanatory variable 204 are expressed as x １ ~x ｐ and the parameter of the reference objective variable 206 is y. In this example, all data from data number A1 to N3 in FIG. 3 are approximated by the following formula 1 through regression analysis. Formula 1 is an example of the estimation model 21 (see FIG. 1). In formula 1, a, b, p, and k are constants.

[0067] The data generating unit 20 calculates the parameter x when y in Equation 1 is the target value Vt. １ ~x ｐ The calculated parameter x １ ~x ｐ The explanatory variable specified by the combination of the following may be the candidate Ve. １ ~x ｐ Therefore, the data generating unit 20 generates a parameter x １ ~x ｐAlthough Equation 1 is for a linear function, all of the data from data numbers A1 to N3 in FIG. 3 may be approximated by a q-order function (q is an integer equal to or greater than 2). All of the data from data numbers A1 to N3 may be approximated by a machine learning model such as a Support Vector Machine (SVM), Gaussian process regression, or Neural Network (NN).

[0068] The data generation unit 20 may select one of the candidates Ve as the optimized data Dt based on the distance d between the generated candidate Ve and the data space 212. The data generation unit 20 may calculate the distance d (e.g., d1 or d2 in FIG. 6 ) between each candidate Ve and a specific data space 212. When the target value Vt is input by one data provider 200, the specific data space 212 is, for example, the data space 212 of the one data provider 200. The data generation unit 20 may calculate the distances (e.g., d1 and d2 in FIG. 6 ) between each candidate Ve and a plurality of data spaces 212.

[0069] The distance between the candidate Ve and one individual explanatory variable 203 (one of the individual explanatory variables 203-1 to 203-n) in one data space 212 is defined as distance d'. The p types of parameters of the individual explanatory variables 203 in one data space 212 (for example, data space 212-1) are defined as x １ ~x ｐ and the parameter x calculated by Equation 1 １ ~x ｐ and the parameter x １ '~x ｐ Then, the distance d' is expressed by the following equation 2.

[0070] A distance d′ may be defined for the individual explanatory variables 203 in each of the multiple data spaces 212 (data spaces 212-1 to 212-n). The data generating unit 20 may calculate the distance d′ for the individual explanatory variables 203 in each of the multiple data spaces 212.

[0071] The data generation unit 20 may calculate the distance d from multiple distances d' between the candidate Ve and each of the multiple individual explanatory variables 203 in the data space 212. The distance d may be the average value, median, or minimum value of the multiple distances d'. The data generation unit 20 may calculate the distance d using a machine learning model that estimates the data space 212. For example, in the k-nearest neighbor method, the data generation unit 20 calculates the distance d as the average value of k distances d' between the candidate Ve and each of the multiple individual explanatory variables 203 in the data space 212. The data generation unit 20 may select the k distances d' in ascending order. The distance d' may be the Euclidean distance or the Mahalanobis distance between the candidate Ve and the individual explanatory variables 203 in the data space 212. The data generation unit 20 may select, from one or more candidates Ve, a candidate Ve whose distance d from a specific data space 212 satisfies a predetermined condition as the optimized data Dt. For example, when a target value Vt is input by one data provider 200, the data generation unit 20 sets the candidate Ve that has the longest distance d from the data space 212 of the one data provider 200 as the optimized data Dt. The white star and black star in Fig. 6 are examples of the candidate Ve of the optimized data Dt, and the black star is an example of the optimized data Dt.

[0072] FIG. 7 is a diagram illustrating an example of learning by the data generating unit 20. As described above, the data generating unit 20 generates generated data 220 including generated explanatory variables 224 and generated objective variables 226 based on the reference data 201. For example, the data generating unit 20 performs machine learning on first reference data 210. The data generating unit 20 may generate a machine learning model using the first reference data 210. The data generating unit 20 may perform machine learning on the relationship between the reference explanatory variables 204 and the reference objective variable 206. The data generating unit 20 may generate a machine learning model that approximates the relationship between the reference explanatory variables 204 and the reference objective variable 206. The data generating unit 20 may generate a first inference model 22 by performing machine learning on the first reference data 210. The first inference model 22 is an example of the estimation model 21 (see FIG. 1 ).

[0073] FIG. 8 is a diagram illustrating an example of inference using the first inference model 22. The data generation unit 20 includes the first inference model 22. When an input explanatory variable is input, the first inference model 22 outputs a first predicted value Vp1 of the generation objective variable 226 corresponding to the input explanatory variable. The input explanatory variable may be a candidate Ve, or may be an explanatory variable other than the candidate Ve provided by one of the data providers 200. The input explanatory variable is input to the first inference model 22, which has undergone machine learning. The first inference model 22 performs machine learning on the relationship between the reference explanatory variable 204 and the reference objective variable 206. Therefore, when the input explanatory variable is input to the first inference model 22, a first predicted value Vp1 of the generation objective variable 226 can be inferred. When a target value Vt is input to the data generation unit 20, the data generation unit 20 outputs optimization data Dt, which is the generation explanatory variable 224 corresponding to the target value Vt.

[0074] The data generating unit 20 may generate one or more candidates Ve of the optimized data Dt in response to an input of the target value Vt. When the target value Vt is input, the data generating unit 20 may generate one or more candidates Ve of the optimized data Dt corresponding to the target value Vt. The first inference model 22 may output a first predicted value Vp1 of the generation objective variable 226 corresponding to the generated candidate Ve. When a candidate Ve is input, the first inference model 22 outputs the corresponding first predicted value Vp1. The first predicted value Vp1 and the target value Vt do not necessarily coincide. The data generating unit 20 may output one of the candidates Ve as the optimized data Dt based on the first predicted value Vp1 and the target value Vt.

[0075] The data generating unit 20 may recommend the optimized data Dt based on the difference between the first predicted value Vp1 and the target value Vt. When the difference between the first predicted value Vp1 and the target value Vt is less than a predetermined threshold, the data generating unit 20 may recommend the candidate Ve corresponding to the first predicted value Vp1 as the optimized data Dt. When the difference between the first predicted value Vp1 and the target value Vt is equal to or greater than a predetermined threshold, the data generating unit 20 may not generate the optimized data Dt. When the first predicted value Vp1 exceeds the target value Vt, the data generating unit 20 may generate the optimized data Dt. When the target value Vt has multiple parameters, the data generating unit 20 may recommend at least one of the Pareto optimal solutions for the multiple parameters as the optimized data Dt. The Pareto optimal solution is a solution that is not dominated by any candidate Ve other than the Pareto optimal solution.

[0076] The first inference model 22 may output, as the most likely predicted value, the first predicted value Vp1 that is closest to the target value Vt from among the multiple first predicted values ​​Vp1 corresponding to the multiple candidates Ve. The data generating unit 20 may set, as the optimized data Dt, the candidate Ve for which the first inference model 22 has output the most likely predicted value from among the candidates Ve input to the first inference model 22.

[0077] The data output unit 40 (see FIG. 4 ) generates output data 228 including the generation explanatory variables 224 based on the generation objective variable 226 and predetermined output conditions Ca, and outputs the generated output data 228. For example, the data output unit 40 outputs the output data 228 including the generation objective variable 226 and the generation explanatory variables 224 corresponding to the generation objective variable 226 to a user 260 who wishes to acquire a generation objective variable 226 of the same type as the generation objective variable 226 generated by the data generation unit 20. For example, the data output unit 40 outputs the output data 228 to a user 260 who wishes to acquire a generation objective variable 226 whose value is in a similar range to the value of the generation objective variable 226 generated by the data generation unit 20. For example, the data output unit 40 outputs the output data 228 if the reward offered by the user 260 is equal to or greater than the reward desired by the data provider 200.

[0078] The output condition Ca may include at least one of the type of the generation object variable 226, the value of the generation object variable 226, and a remuneration to be paid to at least one of the multiple data providers 200 in the case of acquisition. The remuneration is an index representing the value of the generation object variable 226. The index may be, for example, a monetary amount. The data output unit 40 (see FIG. 4 ) may output the output data 228 to a user 260 who wishes to acquire the generation object variable 226, if the amount of remuneration that the user 260 wishes to pay for acquiring the generation object variable 226 satisfies the amount of remuneration included in the output condition Ca.

[0079] There may be multiple users 260 who wish to obtain the generation objective variable 226 generated by the data generation unit 20. The data output unit 40 (see FIG. 4) may output the output data 228 to one of the multiple users 260 who has presented a user condition 262 that satisfies the output condition Ca. If the user conditions 262 of the multiple users 260 satisfy the output condition Ca, the data output unit 40 may output the output data 228 to the user 260 who has presented the user condition 262 that is most advantageous to the data provider 200. The most advantageous user condition 262 is, for example, the highest desired remuneration amount. The user 260 who receives the output data 228 is the so-called successful bidder.

[0080] The data output unit 40 (see FIG. 4 ) may output the output data 228 to the intermediation unit 420 of the user 260 who presented the user condition 262 that satisfies the output condition Ca. The terminal 360 of the user 260 may notify the user 260 that the output data 228 has been output to the intermediation unit 420. The user 260 who has received the notification can obtain the output data 228 by accessing the intermediation unit 420 via the terminal 360.

[0081] The data output unit 40 (see FIG. 4) may output the generation objective variables 226 generated by the data generation unit 20 to multiple users 260 (users 260-1 to 260-n). That is, the data output unit 40 may auction off the generation objective variables 226. The users 260 may receive the output of the generation objective variables 226 and present user conditions 262.

[0082] The data output unit 40 (see FIG. 4 ) may output the generation explanatory variables 224 to at least one user 260 based on the user conditions 262 and the output condition Ca presented by each of the multiple users 260, in order to obtain the generation explanatory variables 224 corresponding to the output generation objective variable 226. For example, the data output unit 40 may output the generation explanatory variables 224 to a user 260 who presented a user condition 262 that satisfies the output condition Ca. The data output unit 40 may not output the generation explanatory variables 224 to a user 260 who presented a user condition 262 that does not satisfy the output condition Ca. The data output unit 40 may notify a user 260 who presented a user condition 262 that does not satisfy the output condition Ca that they did not win the bid for the generation objective variable 226. The user of the information processing device 100 may be the organizer (auctioneer) of the auction.

[0083] When the data output unit 40 (see FIG. 4 ) outputs a generation explanatory variable 224 to at least one user 260 whose user conditions 262 satisfy the output condition Ca, the data output unit 40 may not output the generation objective variable 226 corresponding to the output generation explanatory variable 224 to other users 260 for a predetermined period. The predetermined period may be a period for which the at least one user 260 desires to monopolize the generation objective variable 226, as specified in the user conditions 262 of the at least one user 260 to whom the generation explanatory variable 224 was output. The generation objective variable 226 may be excluded from the auction for that period. As a result, the user 260 who wins the bid for the generation objective variable 226 can monopolize the generation explanatory variable 224 corresponding to the generation objective variable 226. The period for which the user 260 monopolizes the generation explanatory variable 224 may be, for example, one week, two weeks, or one month.

[0084] When the data output unit 40 (see FIG. 4 ) outputs a generation explanatory variable 224 to at least one user 260 whose user conditions 262 satisfy the output condition Ca, the data output unit 40 does not have to output the output generation explanatory variable 224 and other generation explanatory variables that are within a predetermined range of similarity to the output generation explanatory variable 224 to other users 260 for a predetermined period of time. This allows the user 260 who wins the bid for the generation objective variable 226 to monopolize the generation explanatory variable 224 that corresponds to the generation objective variable 226 and the other generation explanatory variables 224 that are within a similarity range to the output generation explanatory variable 224.

[0085] The data generator 20 may determine the range of similarity of one generation explanatory variable 224 to another generation explanatory variable 224 based on the similarity between the one generation explanatory variable 224 and the other generation explanatory variable 224. The similarity between one generation explanatory variable 224 and the other generation explanatory variable 224 is, for example, the distance between the one generation explanatory variable 224 and the other generation explanatory variable 224. This distance is referred to as distance d1.

[0086] In the example of FIG. 3, p types of parameters of the individual explanatory variables 203 are expressed as x １１ ~x １ｐ When the individual explanatory variables 203 have p types of parameters, the generating explanatory variables 224 have p types of parameters. The p types of parameters of one generating explanatory variable 224 are expressed as x １１ ~x １ｐ and the other p types of parameters of the generation explanatory variables 224 are x ２１ ~x ２ｐ Then, the distance d1 is expressed by the following equation 3. In the example of FIG.

[0087] The data generating unit 20 may calculate the distance d1 using Equation 3. The data generating unit 20 may calculate the distance d1 for each of all combinations of two generation explanatory variables 224 selected from the generated plurality of generation explanatory variables 224. The data generating unit 20 may calculate the threshold distance d1 based on the calculated plurality of distances d1. ｔｈ For example, the data generating unit 20 may calculate the average value, median value, or minimum value of the plurality of distances d1 as the threshold distance d1. ｔｈThe data generating unit 20 calculates the distance d1 between one generation explanatory variable 224 and another generation explanatory variable 224 as a threshold distance d1 ｔｈ It may be determined that the one generating explanatory variable 224 and the other generating explanatory variable 224 are in a similar range if:

[0088] The data generating unit 20 may determine the similarity range of one generation explanatory variable 224 with respect to the other generation explanatory variables 224 based on the similarity between one generation explanatory variable 224 and the other generation explanatory variables 224 and the similarity between one generation objective variable 226 and the other generation objective variables 226. The similarity between one generation objective variable 226 and the other generation objective variables 226 is, for example, the distance between the one generation objective variable 226 and the other generation objective variables 226. This distance is referred to as distance d2.

[0089] In the example of FIG. 3, r types of parameters of the individual objective variables 205 are expressed as y １１ ~y １ｒ When the individual objective variable 205 has r types of parameters, the generation objective variable 226 has r types of parameters. The r types of parameters of one generation objective variable 226 are expressed as y １１ ~y １ｒ and the other r types of parameters of the generated objective variable 226 are y ２１ ~y ２ｒ Then, the distance d2 is expressed by the following equation 4. In the example of FIG.

[0090] The data generating unit 20 may calculate the distance d2 using Equation 4. The data generating unit 20 may calculate the distance d2 for each of all combinations of two generation objective variables 226 selected from the generated plurality of generation objective variables 226. The data generating unit 20 may calculate the threshold distance d2 based on the calculated plurality of distances d2. ｔｈ For example, the data generating unit 20 may calculate the average value, median value, or minimum value of the plurality of distances d2 as the threshold distance d2 ｔｈ It is calculated as follows.

[0091] The data generating unit 20 determines whether the distance d1 between one generation explanatory variable 224 and another generation explanatory variable 224 is a threshold distance d1 ｔｈis less than or equal to the threshold distance d2, and the distance d2 between one generation objective variable 226 corresponding to the one generation explanatory variable 224 and another generation objective variable 226 corresponding to the other generation explanatory variable 224 is less than or equal to the threshold distance d2 ｔｈ In the following cases, it may be determined that the one generation explanatory variable 224 and the other generation explanatory variable 224 are in a similarity range. As a result, the user 260 who wins the bid for the generation objective variable 226 can monopolize not only the generation explanatory variable 224 corresponding to the generation objective variable 226 and the other generation explanatory variables 224 that are in a similarity range to the generation explanatory variable 224, but also the other generation objective variables 226 that are in a similarity range to the generation objective variable 226.

[0092] The storage unit 50 (see FIG. 4 ) may store the user conditions 262. When the user conditions 262 of at least one user 260 satisfy the output condition Ca for a generation objective variable 226 that at least one user 260 wishes to acquire, the data output unit 40 may output the generation explanatory variable 224 corresponding to the generation objective variable 226 to the at least one user 260.

[0093] The user conditions 262 may include the generation objective variables 226 that each of the multiple users 260 wishes to acquire, and acquisition conditions Cb for acquiring the generation objective variables 226. The acquisition conditions Cb may include multiple acquisition parameters. The acquisition parameters may include, for example, the type of generation objective variable 226 that the user 260 wishes to acquire, the value of the generation objective variable 226, the range of the value of the generation objective variable 226, the desired monopoly period for which the user 260 wishes to monopolize the generation objective variable 226, and the reward that the user 260 will pay for acquiring the generation objective variable 226. The types and values ​​of the acquisition parameters may be provided by the user 260.

[0094] The output condition Ca may include a plurality of output parameters. The output parameters may include, for example, the type of outputtable generation objective variable 226, the value of the generation objective variable 226, the range of values ​​of the generation objective variable 226, the period during which the generation objective variable 226 can be exclusively used, and the remuneration desired by the data provider 200 as consideration for the transfer of the generation objective variable 226 and the generation explanatory variable 224. The types and values ​​of the output parameters may be provided by the data provider 200.

[0095] The data output unit 40 may read out the user conditions 262 stored in the storage unit 50 and determine whether the read out user conditions 262 satisfy the output conditions Ca. If the data output unit 40 determines that the user conditions 262 satisfy the output conditions Ca, it may output the generation explanatory variables 224. This allows the data output unit 40 to determine whether the conditions are desired by the user 260 and the data provider 200, and to output the generation explanatory variables 224 if the conditions are desired.

[0096] When the data output unit 40 determines that at least one condition of the multiple acquisition parameters in the acquisition conditions Cb satisfies at least one condition of the multiple output parameters in the output conditions Ca, it may output the generation explanatory variable 224. For example, when the value range of the generation objective variable 226 in the acquisition parameters is included in the value range of the generation objective variable 226 in the output parameters, the data output unit 40 may output the generation explanatory variable 224. When the value range of the generation objective variable 226 in the acquisition parameters is included in the value range of the generation objective variable 226 in the output parameters and the reward in the acquisition parameters satisfies the reward in the output parameters, the data output unit 40 may output the generation explanatory variable 224.

[0097] Priorities may be assigned to the plurality of acquisition parameters and the plurality of output parameters, respectively. The priorities may be stored in the storage unit 50. The priorities of the plurality of acquisition parameters may be provided by the user 260. The priorities of the plurality of output parameters may be provided by the data provider 200.

[0098] When determining whether the user condition 262 satisfies the output condition Ca, the data output unit 40 may read the priority order stored in the storage unit 50 and determine whether the acquisition parameters satisfy the output parameters in descending order of priority of the acquisition parameters, or may determine whether the acquisition parameters satisfy the output parameters in descending order of priority of the output parameters. In this way, the data output unit 40 can determine whether the conditions are desired by the user 260 and the data provider 200 in accordance with the priority order, and can output the generation explanatory variable 224 if the desired conditions are met.

[0099] The reward paid by the user 260, which is included in the acquisition parameters, may be monetary. The user condition 262 may include the amount paid by the user 260. When the user condition 262 of each of the multiple users 260 satisfies the output condition Ca, the data output section 40 may determine one user 260 who will output the generation explanatory variable 224 based on the amount paid by the user 260. For example, the data output section 40 determines the user 260 who offered the highest amount among the amounts offered by each of the multiple users 260 as the user 260 who will output the generation explanatory variable 224.

[0100] The data provider 200 may provide a desired amount of money as consideration for the transfer of the generation objective variable 226 and the generation explanatory variable 224. When the data output section 40 determines that all of the amounts offered by two or more users 260 among the multiple users 260 satisfy the amount offered by the data provider 200, the data output section 40 may determine the user 260 who has offered the widest range of values ​​for the generation objective variable 226 among the ranges offered by each of the multiple users 260 as the user 260 to output the generation explanatory variable 224, or may determine the user 260 who has offered the longest period of time among the desired periods for monopolizing the generation objective variable 226 among the desired periods of time offered by each of the multiple users 260 as the user 260 to output the generation explanatory variable 224.

[0101] The data output unit 40 may conceal the user conditions 262 of one user 260 from the other users 260. For example, the data output unit 40 conceals the user conditions 262-1 of user 260-1 from users 260-2 to 260-n'. The data output unit 40 may conceal the user conditions 262 of one user 260 from the other users 260, for each of the user conditions 262 of multiple users 260 stored in the storage unit 50. The user conditions 262 are highly likely to involve confidential matters of the user 260. For this reason, it is highly likely that the one user 260 will want to conceal the user conditions 262 that he or she presents from the other users 260. By the data output unit 40 concealing the user conditions 262 of the one user 260 from the other users 260, the one user 260 can participate in the auction of the generation target variable 226 with peace of mind.

[0102] When the data output unit 40 decides to output the generation explanatory variables 224 to one user 260, it may conceal the user conditions 262 of that one user 260 from the other users 260. The data output unit 40 does not need to conceal the user conditions 262 of that one user 260 from the other users 260 before deciding to output the generation explanatory variables 224 to that one user 260. When a user 260 wins a bid for the generation objective variable 226, it is highly likely that the user 260 will want to conceal the user conditions 262 related to the successful bid for the generation objective variable 226 from the other users 260. For this reason, when the data output unit 40 decides to output the generation explanatory variables 224 to one user 260, it may conceal the user conditions 262 of that one user 260 from the other users 260.

[0103] The data output unit 40 may output information indicating other generation objective variables 226 that are recommended for acquisition to one user 260, based on the generation objective variable 226 corresponding to the generation explanatory variable 224 output to the one user 260 of the generation data 220. For example, the data output unit 40 may recommend to the one user 260 that the other generation objective variables 226 be acquired within a predetermined similarity range for the generation objective variable 226 corresponding to the output generation explanatory variable 224. The similarity range is determined when the distance d2 between the generation objective variable 226 corresponding to the generation explanatory variable 224 output by the data output unit 40 and the other generation objective variable 226 is equal to or smaller than the above-mentioned threshold distance d2. ｔｈ The information indicating the other generation objective variable 226 whose acquisition is recommended may include a threshold distance d2 ｔｈ may include information on

[0104] For one user 260, one exclusive range regarding the generation objective variable 226 may be determined in advance. The one exclusive range may be presented by the one user 260. The one user 260 may register the one exclusive range in the intermediation unit 420 (see FIG. 1 ). For another user 260, another exclusive range regarding the generation objective variable 226 may be determined in advance. The other exclusive range may be presented by the other user 260. The other user 260 may register the other exclusive range in the intermediation unit 420.

[0105] "One exclusive range is determined in advance" means that the exclusive range is determined before one user 260 wins the bid for the generation objective variable 226. The exclusive range may be the exclusive range that the one user 260 desires for the generation objective variable 226 if the one user 260 wins the bid for the generation objective variable 226. The above also applies to the other exclusive ranges described above.

[0106] If the generation objective variable 226 that one user 260 wishes to acquire belongs to another exclusive range predetermined for another user 260, the data output unit 40 does not need to output the generation objective variable 226 that belongs to the other exclusive range to the one user 260. The data output unit 40 may notify the one user 260 that the generation objective variable 226 that the one user 260 wishes to acquire belongs to the other exclusive range predetermined for the other user 260. This allows the one user 260 to change the generation objective variable 226 that the one user 260 wishes to acquire without waiting for an auction. This also eliminates the concern of the other user 260 that the generation objective variable 226 that belongs to the user's exclusive range (the other exclusive range) will be won by a user 260 different from the one user 260.

[0107] If the generation objective variable 226 that one user 260 wishes to acquire belongs to another exclusive range predetermined for another user 260, the data output unit 40 may determine that the generation objective variable 226 that belongs to the other exclusive range does not satisfy the output condition Ca. For example, even if the generation objective variable 226 that one user 260 wishes to acquire belongs to another exclusive range predetermined for the other user 260, the data output unit 40 outputs the generation objective variable 226 that belongs to the other exclusive range to the one user 260, but determines that the generation objective variable 226 does not satisfy the output condition Ca. This prevents the one user 260 from recognizing that he or she was unsuccessful in the bid because the generation objective variable 226 belongs to the other exclusive range. Therefore, the other user 260 can easily win the bid for the generation objective variable 226 that belongs to his or her own exclusive range (the other exclusive range) without the other user 260 knowing about his or her own exclusive range (the other exclusive range).

[0108] When one exclusive range predetermined for one user 260-1 overlaps with another exclusive range predetermined for another user 260-2, the data output unit 40 may output the generation explanatory variable 224 to either the one user 260-1 or the other user 260-2 based on the user conditions 262-1 of the one user 260-1 and the user conditions 262-2 of the other user 260-2. That is, the data output unit 40 may determine either the one user 260-1 or the other user 260-2 as the successful bidder for the generation objective variable 226. For example, the data output unit 40 compares the remuneration paid by the one user 260-1 to the data provider 200 with the remuneration paid by the one user 260-2 to the data provider 200, determines the user 260 who pays the higher remuneration as the successful bidder, and outputs the generation explanatory variable 224 to the determined user 260. The user conditions 262 may include a reward that the user 260 will pay to the data provider 200 if the user 260 wins the bid for the generation target variable 226 .

[0109] When the data output unit 40 outputs the generation explanatory variables 224 to the user 260 of the generation data 220, it may output a reward to the data provider 200 that provided the reference explanatory variables 204, based on the distance between the output generation explanatory variables 224 and the reference explanatory variables 204. As described above, the reference explanatory variables 204 provided by one data provider 200 are the individual explanatory variables 203. Therefore, when the data output unit 40 outputs the generation explanatory variables 224 to the user 260, it may output a reward to the at least one data provider 200, based on the distance between the output generation explanatory variables 224 and the individual explanatory variables 203 provided by the at least one data provider 200.

[0110] The p types of parameters of the individual explanatory variables 203 are expressed as x １１ ~x １ｐ The p types of parameters in the generation explanatory variables 224 output by the data output unit 40 are expressed as x ２１ ~x ２ｐ Then, the distance between the output generation explanatory variable 224 and the individual explanatory variable 203 provided by one data provider 200 is defined as the distance d1 in Equation 3. The data generating unit 20 may calculate the distance d1 using Equation 3.

[0111] For example, the data output unit 40 outputs a reward according to the distance d1 to the data provider 200. For example, the shorter the distance d1, the higher the reward the data output unit 40 outputs to the data provider 200, and the longer the distance d1, the lower the reward the data output unit 40 outputs to the data provider 200. The data output unit 40 may output rewards to multiple data providers 200 according to the distance d1. The data output unit 40 may output a reward to the data provider 200 that provided the individual explanatory variable 203 with the smallest distance d1, among the multiple data providers 200. The reward is, for example, money, as described in the description of FIG. 1 .

[0112] When a single data provider 200 provides multiple sets of individual explanatory variables 203 and individual objective variables 205, the data output unit 40 does not need to output to the single data provider 200 which of the multiple sets the individual explanatory variable 203 with the smallest distance d1 belongs to. When a reward is output to the single data provider 200, the generated explanatory variable 224 that is separated by the distance d1 from the individual explanatory variable 203 provided by the single data provider 200 has been won by at least one of the multiple users 260. Therefore, when the data output unit 40 outputs to the single data provider 200 which of the multiple sets the individual explanatory variable 203 with the smallest distance d1 belongs to, the single data provider 200 will recognize the individual explanatory variable 203 with the smallest distance d1. Therefore, the single data provider 200 will be more likely to provide other individual explanatory variables 203 that may be won by other users 260, without providing individual explanatory variables 203 close to the single data provider 203. This may result in bias in the types, values, and ranges of the individual explanatory variables 203. If such bias occurs in the individual explanatory variables 203, the accuracy of the estimation model 21 (see FIG. 1 ) may decrease. For this reason, the data output unit 40 does not need to output to the one data provider 200 to which of the multiple sets the individual explanatory variable 203 with the smallest distance d1 belongs.

[0113] 8 is a diagram showing another example of learning by the data generating unit 20. The second reference data 250 is reference data obtained by excluding, from the first reference data 210, the individual explanatory variables 203 and the individual objective variables 205 corresponding to at least one data provider 200 among the multiple data providers 200. The second reference data 250 is, for example, reference data obtained by excluding, from the first reference data 210, the individual explanatory variables 203-1 and the individual objective variables 205-1 of the data provider 200-1 (Company A, see FIG. 3), and is reference data into which the individual explanatory variables 203 and the individual objective variables 205 of the data providers 200-2 to 200-n (Company B to Company N, see FIG. 3) are aggregated.

[0114] The data generating unit 20 may perform machine learning on the second reference data 250. The data generating unit 20 may generate a machine learning model using the second reference data 250. The data generating unit 20 performs machine learning on the relationship between the individual explanatory variables 203 and the individual objective variable 205. The data generating unit 20 may generate a machine learning model that approximates the relationship between the individual explanatory variables 203 and the individual objective variable 205. The data generating unit 20 may generate a second inference model 24 by performing machine learning on the second reference data 250. The data generating unit 20 may include the second inference model 24.

[0115] When an input explanatory variable is input, the second inference model 24 outputs a second predicted value Vp2 of the generation target variable 226 corresponding to the input explanatory variable. The contribution calculation unit 60 (see FIG. 4) may calculate the contribution of one first reference data 210 corresponding to one data provider 200 to the optimized data Dt based on the first predicted value Vp1 and the second predicted value Vp2. The one first reference data 210 corresponding to one data provider 200 is, in other words, the individual reference data 202 (see FIG. 3) of the one data provider 200.

[0116] When the second inference model 24 is generated by machine learning the second reference data 250 from which one individual reference data 202-u (1≦u≦n) is excluded, the second predicted value Vp2 output by the second inference model 24 is Vp2 ｕFor example, the contribution calculation unit 60 calculates the contribution C of one reference teacher data 202-u to the optimized data Dt using the following formula 4: u Calculate.

[0117] As shown in Equation 5, the contribution calculation unit 60 may calculate the contribution to the optimized data Dt based on the difference between the first predicted value Vp1 and the second predicted value Vp2. The contribution calculation unit 60 may evaluate the contribution to the optimized data Dt as greater the greater the difference.

[0118] When the difference between the first predicted value Vp1 and the second predicted value Vp2 is relatively small, the impact on the first predicted value Vp1 of excluding one individual reference data 202 from the first reference data 210 (see FIG. 3 ) is relatively small. Therefore, the contribution calculation unit 60 may calculate the contribution of the one individual reference data 202 to the optimized data Dt to be smaller as the difference between the first predicted value Vp1 and the second predicted value Vp2 is smaller.

[0119] When the reference objective variable 206 and the individual objective variable 205 (see FIG. 3 ) have multiple types of parameters, the contribution calculation unit 60 may calculate the contribution of the individual reference data 202 to the optimized data Dt for each of the multiple types of parameters. The contribution calculation unit 60 may calculate the average, median, minimum, or maximum of the contribution calculated for each of the multiple types of parameters as the contribution of the individual reference data 202 to the optimized data Dt.

[0120] 3 , the reference objective variable 206 and the individual objective variable 205 have two types of parameters. In the example of FIG. 3 , the contribution calculation unit 60 may calculate the contribution of the individual reference data 202 to the optimized data Dt for each of the two types of parameters ("stiffness" and "extensibility"). The contribution calculation unit 60 may calculate the average, median, minimum, or maximum of the contribution calculated for one parameter (stiffness) and the contribution calculated for the other parameter (extensibility) as the contribution of the individual reference data 202 to the optimized data Dt.

[0121] When the optimized data Dt is input, the first inference model 22 may output a first predicted value Vp1 corresponding to the optimized data Dt. The first predicted value Vp1 corresponding to the optimized data Dt is defined as a first predicted value Vp1'. When the optimized data Dt is input, the second inference model 24 may output a second predicted value Vp2 corresponding to the optimized data Dt. The second predicted value Vp2 corresponding to the optimized data Dt is defined as a second predicted value Vp2'.

[0122] The contribution calculation unit 60 may calculate the contribution of the individual reference data 202 to the optimized data Dt based on the first predicted value Vp1' and the second predicted value Vp2'. For example, the contribution calculation unit 60 calculates the contribution of the individual reference data 202 to the optimized data Dt based on the difference between the first predicted value Vp1' and the second predicted value Vp2'. The contribution calculation unit 60 may evaluate the contribution of the individual reference data 202 to the optimized data Dt to be higher the larger the difference between the first predicted value Vp1' and the second predicted value Vp2'.

[0123] 9 is a diagram showing another example of inference using the second inference model 24. The data generation unit 20 may perform machine learning on the second reference data 250 for each of the multiple data providers 200, from which the individual explanatory variables 203 and the individual objective variables 205 corresponding to one of the multiple data providers 200 have been excluded. In the example of FIG. 9 , the data generation unit 20 performs machine learning on the second reference data 250-u for each of n data providers 200, from which the individual explanatory variables 203-u and the individual objective variables 205-u corresponding to the data provider 200-u (1≦u≦n) have been excluded, thereby generating second inference models 24 (second inference models 24-1 to 24-n) for each of the multiple data providers 200.

[0124] 10 is a diagram showing another example of inference by the second inference model 24. The second inference model 24 may output a second predicted value Vp2 for each of multiple data providers 200. In the example of FIG. 10, the second inference model 24-1 outputs second predicted values ​​Vp2-1 to Vp2-n for data providers 200-1 to 200-n, respectively.

[0125] The contribution calculation unit 60 may calculate the contribution of each of the multiple data providers 200 to the optimized data Dt based on the first predicted value Vp1 and the second predicted value Vp2 of each of the multiple data providers 200. In the example of Fig. 10, when n is 2, the contribution calculation unit 60 calculates the contribution of the data provider 200-1 based on the first predicted value Vp1 and the second predicted value Vp2-1, and calculates the contribution of the data provider 200-2 based on the first predicted value Vp1 and the second predicted value Vp2-2.

[0126] For example, the contribution calculation unit 60 calculates the difference between the first predicted value Vp1 and the second predicted value Vp2 of each of the multiple data providers 200, and evaluates the contribution of the data provider 200 to the optimized data Dt higher the larger the calculated difference. The contribution calculation unit 60 may rank the contributions of the data providers 200 based on the magnitude of the difference between the first predicted value Vp1 and the second predicted value Vp2 of each of the multiple data providers 200.

[0127] When the data output unit 40 outputs the generation explanatory variable 224 to the user 260 (see FIG. 1 ) of the generation data 220, the data output unit 40 may output a reward to each of the multiple data providers 200 in accordance with the degree of contribution of each of the multiple data providers 200 to the output generation explanatory variable 224. For example, the data output unit 40 outputs a larger reward to a data provider 200 whose degree of contribution to the output generation explanatory variable 224 is greater. When the data output unit 40 outputs the generation explanatory variable 224 to the user 260 (i.e., when the user 260 wins the bid for the generation explanatory variable 224), each of the multiple data providers 200 contributed to the generation of the generation explanatory variable 224. Therefore, a reward may be output to each of the multiple data providers 200. A reward may be output to each of the multiple data providers 200 in accordance with their respective degrees of contribution.

[0128] 11 is a diagram showing another example of learning by the data generating unit 20. The third reference data 230 is reference data obtained by excluding one reference explanatory variable 204 and one reference objective variable 206 corresponding to the one reference explanatory variable from the first reference data 210.

[0129] The number of sets in the first reference data 210, each consisting of an individual explanatory variable 203 and an individual objective variable 205 corresponding to the individual explanatory variable, is defined as the number of sets Se. In the example of FIG. 3, the number of sets Se is 3n. The number of sets Se in the first reference data 210 is, for example, 100. The 100 sets in the first reference data 210 are defined as the first set Se1 to the hundredth set Se100. The third reference data 230 is, for example, reference data obtained by excluding the first set Se1 to the tenth set Se10 from the first reference data 210, and is reference data obtained by aggregating the eleventh set Se11 to the hundredth set Se100. In this example, the individual explanatory variables 203 of the first set Se1 to the tenth set Se10 correspond to the above-mentioned one reference explanatory variable 204, and the individual objective variables 205 of the first set Se1 to the tenth set Se10 correspond to the above-mentioned one reference objective variable 206. In this example, the reference data of the first set Se1 to the tenth set Se10, which are the reference data to be excluded, do not need to correspond to one data provider 200. The reference data may include individual reference data 202 provided by multiple data providers 200.

[0130] The data generating unit 20 may perform machine learning on the third reference data 230. The data generating unit 20 may generate a third inference model 26 by performing machine learning on the third reference data 230. The data generating unit 20 may include the third inference model 26. When an input explanatory variable is input, the third inference model 26 outputs a third predicted value Vp3 of the generation target variable 226 corresponding to the input explanatory variable.

[0131] The reliability evaluation unit 70 (see FIG. 4 ) evaluates the reliability of the first reference data 210 based on the first predicted value Vp1 and the third predicted value Vp3. For example, the reliability evaluation unit 70 evaluates the reliability of the first reference data 210 based on the difference between the first predicted value Vp1 and the third predicted value Vp3.

[0132] The multiple users 260 may predetermine one or more thresholds of reliability. The one or more thresholds may be determined by consensus of the multiple users 260. The determined thresholds may be stored in the storage unit 50 (see FIG. 4 ). The thresholds may be included in the user conditions 262.

[0133] The reliability of the first reference data 210 may be divided into multiple levels in advance. The multiple levels are, for example, three levels represented by high, medium, and low reliability. When there are three levels, a first threshold for determining whether the reliability is high or medium and a second threshold for determining whether the reliability is medium or low may be determined in advance. In this example, the second threshold is greater than the first threshold. The number of levels may be determined by agreement among multiple users 260. The number of levels may be stored in the storage unit 50 (see FIG. 4). The number of levels may be included in the user conditions 262.

[0134] The reliability evaluation unit 70 (see FIG. 4 ) may evaluate whether the reliability of the first reference data 210 is in one of a plurality of levels based on the difference between the first predicted value Vp1 and the third predicted value Vp3 and one or more reliability thresholds. For example, if the reliability of the first reference data 210 is divided into two levels (high and low), the reliability evaluation unit 70 may determine that the reliability level of the first reference data 210 is high if the difference between the first predicted value Vp1 and the third predicted value Vp3 is smaller than the reliability threshold, and may determine that the reliability level of the first reference data 210 is low if the difference is equal to or greater than the reliability threshold. For example, if the reliability of the first reference data 210 is divided into three levels (high, medium, and low), the reliability evaluation unit 70 may determine that the reliability level of the first reference data 210 is high if the difference between the first predicted value Vp1 and the third predicted value Vp3 is smaller than the first reliability threshold, may determine that the reliability level of the first reference data 210 is medium if the difference is greater than or equal to the first reliability threshold and smaller than the second reliability threshold, and may determine that the reliability level of the first reference data 210 is small if the difference is greater than or equal to the second reliability threshold.

[0135] For example, if the reliability level of the first reference data 210 is low, there is a high probability that anomalous data was included in the data excluded from the first reference data 210 in order to aggregate the third reference data 230. Therefore, the reliability evaluation unit 70 (see FIG. 4) can evaluate that there is a high probability that the first reference data 210 includes anomalous data.

[0136] The data output unit 40 may determine whether or not to output the generation object variable 226 generated by the data generation unit 20 to the user 260, based on the reliability evaluated by the reliability evaluation unit 70. For example, if the reliability evaluated by the reliability evaluation unit 70 is equal to or greater than a reliability threshold, the data output unit 40 outputs the generation object variable 226 to the user 260. That is, the data output unit 40 puts the generation object variable 226 up for auction. This makes it more difficult for the user 260 to obtain a generation object variable 226 with low reliability.

[0137] Each of the multiple users 260 may determine one or more thresholds of reliability that they accept. The thresholds may be different for each of the multiple users 260. The thresholds may be stored in the storage unit 50 (see FIG. 4). In this case, the data output unit 40 may output the generation target variable 226 generated by the data generation unit 20, regardless of the reliability evaluated by the reliability evaluation unit 70. The data output unit 40 may output the reliability together.

[0138] The intermediation unit 420 (see FIG. 1 ) of each of the multiple users 260 may determine whether the reliability output by the data output unit 40 is greater than or equal to one or more thresholds for reliability. If the reliability is equal to or greater than the threshold, the intermediation unit 420 may output the generation objective variable 226 output by the data output unit 40 to the user 260. As a result, if the reliability output by the data output unit 40 is less than the threshold that the intermediation unit 420 itself accepts, the generation objective variable 226 will not be offered for auction to the user 260 who set the threshold.

[0139] When a user 260 (see FIG. 1 ) wishes to acquire new generation explanatory variables 224 corresponding to the generation objective variables 226 acquired by the user 260, the user 260 may transmit a request to the information processing device 100 via the network 400 to have the data output unit 40 (see FIG. 4 ) output the new generation explanatory variables 224. When a request to output the new generation explanatory variables 224 is input to the input unit 30 (see FIG. 4 ), the data generation unit 20 (see FIG. 4 ) may generate new generation data 220. The new generation data 220 may be generated by the estimation model 21 (see FIG. 1 ) without machine learning, or may be generated by a first inference model 22 (see FIG. 7 ) that has undergone machine learning. In this way, the user 260 can acquire the new generation explanatory variables 224.

[0140] When new individual reference data 202 is obtained, each of the multiple data providers 200 may transmit the new individual reference data 202 to the information processing device 100 via the network 400. The new individual reference data 202 includes individual explanatory variables 203 and individual objective variables 205 corresponding to at least one of the multiple data providers 200. When new individual reference data 202 is newly input to the input unit 30 (see FIG. 4), the data generation unit 20 (see FIG. 4) may generate new generated data 220. The new generated data 220 may be generated by the estimation model 21 (see FIG. 1) without machine learning, or may be generated by a first inference model 22 (see FIG. 7) that has undergone machine learning. By generating the new generated data 220, the user 260 may obtain more accurate generated explanatory variables 224.

[0141] FIG. 12 is a diagram showing another example of first reference data 210 provided by each of multiple data providers 200. When the reference explanatory variable 204 has multiple parameters (15 in this example, from "raw material A" to the "reference objective variable 206" of evaluation condition E), at least one parameter may be provided by at least one data provider 200. The individual reference data 208 may be provided by each of the multiple data providers 200. In this example, individual reference data 208-1 to 208-6 are provided by data providers 200-1 to 200-6, respectively, and individual reference data 208-7 is provided by data provider 200-6. In this example, the first reference data 210 is data obtained by aggregating the individual training data 208-1 to 208-7. This example differs from the example in FIG. 3 .

[0142] In this example, the reference explanatory variable 204 is provided by data providers 200-1 to 200-6, and the reference objective variable 206 is provided by data provider 200-6. In this example, the composition Co of the main material Sm is provided by data provider 200-1, the composition Co of the secondary material Sb is provided by data provider 200-2, and the composition Co of the additional material Sa is provided by data provider 200-3. In this example, the process condition P1 is provided by data provider 200-4, the process condition P2 is provided by data provider 200-5, and the evaluation condition Ev and the reference objective variable 206 are provided by data provider 200-6.

[0143] A plurality of types of parameters may be provided by at least one data provider 200. In this example, data provider 200-1 provides three types of parameters, i.e., the composition Co of raw material A to the composition Co of raw material C. In this example, data providers 200-1 to 200-6 similarly provide a plurality of types of parameters, respectively.

[0144] The data generating unit 20 (see FIG. 4 ) may generate the first inference model 22 by machine learning the reference explanatory variables 204 and the reference objective variables 206. When a target value Vt of the objective variable to be generated is input to the data generating unit 20, the data generating unit 20 outputs optimization data Dt, which is an explanatory variable corresponding to the target value Vt.

[0145] The contribution calculation unit 60 may calculate the contribution of at least one parameter to the optimized data Dt based on the optimized data Dt and the first reference data 210. In this example, the contribution calculation unit 60 calculates the contribution of each of the individual reference data 208-1 to 208-6 to the optimized data Dt based on the optimized data Dt and the first reference data 210. The contribution calculation unit 60 may calculate the contribution h of at least one species to the optimized data Dt for each of the multiple species. For example, the contribution calculation unit 60 may calculate the contribution h of each of the multiple species of parameters to the optimized data Dt based on the optimized data Dt and the first reference data 210.

[0146] The contribution degree h may be calculated based on the average predicted value Pa of the first inference model. The predicted value Pa may be the average value of the objective variable in multiple optimization data output by the first inference model, or may be the average value of the objective variable in the first teacher data 210. Specifically, data is generated in which all but some of the explanatory variables x i in the optimization data Dt are replaced with the average value of the explanatory variables included in the first reference data 210, and this data is input to the first inference model to obtain the predicted value Ph of the objective variable. Then, the contribution degree h of x i may be calculated based on the difference between Ph and Pa.

[0147] For simplicity, we will explain an example in which there are three explanatory variables (x1, x2, x3) and one objective variable (y). In this case, the explanatory variables of the optimized data are (x1o, x2o, x3o), the predicted value of the objective variable when the optimized data is input to the first inference model is yo, the average values ​​of the explanatory variables in the first teacher data are (x1a, x2a, x3a), and the average value of the objective variable y is Pa.

[0148] When yo = 50 and Pa = 30, the difference of 20 is the total contribution of the optimized data. Here, when calculating the contribution of each explanatory variable, we will further explain using x1 as an example. Suppose that by inputting (x1o, x2a, x3a) into the first inference model, the predicted value Ph of the objective variable is 35. In this case, the contribution of x1 can be calculated as h = 5, which is the difference between Ph = 35 and Pa = 30.

[0149] Specifically, the Shapley value or its approximate value, the SHAP (Shapley Additive exPlanations) value, may be calculated as the contribution h. The Shapley value h is a distribution algorithm used when distributing rewards earned through the cooperation of multiple data providers according to the respective contributions of the multiple data providers. The contribution calculation unit 60 may calculate the Shapley value h based on the optimized data Dt and the first reference data 210 using a known method (e.g., paragraphs

[0007] to

[0010] of Japanese Patent No. 7086497).

[0150] The contribution calculation unit 60 calculates the contribution h of the p types of parameters of the reference explanatory variables 204 to the optimized data Dt. ｓ (1≦s≦p) for each of the plurality of species. ｓ For example, the contribution calculation unit 60 calculates the contribution C of each of the plurality of individual reference data 208 to the optimized data Dt using the following formula 6:

[0151] In Equation 6, p is the number of all parameters in the reference explanatory variables 204 (15 in this example). The denominator of Equation 6 is the sum of the contributions of all parameters in the reference explanatory variables 204. In Equation 6, i is the number of parameters provided by each data provider 200. For example, i for data provider 200-1 is 3. The numerator of Equation 6 is the sum of the contributions of parameters in one or more reference explanatory variables 204 provided by each data provider 200.

[0152] 13 is a flowchart illustrating an example of an information processing method according to an embodiment of the present invention. The information processing method according to an embodiment of the present invention will be described using the information processing device 100 illustrated in FIG. 4 as an example. The information processing method may include a data providing and aggregating step. The data providing and aggregating step may include a data providing step S70, a determination step S80, a missing parameter adding step S82, a data aggregating step S84, and a preprocessing step S90. The preprocessing step S90 may include a determination step S92 and a missing value adding step S94.

[0153] In the data providing step S70, each of the multiple data providers 200 (data providers 200-1 to 200-n) provides individual reference data 202 (individual reference data 202-1 to 202-n). In the determining step S80, the data generating unit 20 determines whether at least one of the parameters (15 parameters in the example of FIG. 3 , from "raw material A" to "temperature" of the evaluation condition Ev) is missing in at least one of the individual reference data 202-1 to 202-n. If it is determined that at least one of the parameters is missing, the information processing method proceeds to a missing parameter adding step S82. If it is determined that at least one of the parameters is not missing, the information processing method proceeds to a data aggregating step S84.

[0154] The data aggregation step S84 is a step in which the data generation unit 20 aggregates the individual reference data 202-1 to 202-n. In the data aggregation step S84, the data generation unit 20 aggregates the individual reference data 202-1 to 202-n to generate the first reference data 210 (see FIG. 3).

[0155] In the determination step S92, the data generating unit 20 determines whether there are any parameters with missing values ​​in the first reference data 210 (see FIG. 3) generated in the data aggregating step S84. If it is determined that there are any parameters with missing values, the information processing method proceeds to the missing value addition step S94. If it is not determined that there are any parameters with missing values, the information processing method ends the data providing and aggregating step.

[0156] 14 is a flowchart showing an example of an information processing method according to an embodiment of the present invention. The information processing method includes a specifying step S100. The specifying step may include a spatial information saving step S102 and a determination step S104.

[0157] The identification step S100 is a step in which the identification unit 10 identifies the data spaces 212 of each of the multiple data providers 200 based on the individual reference data 202 having reference explanatory variables 204 provided by each of the multiple data providers 200. The spatial information saving step S102 is a step in which the data generation unit 20 saves the spatial information of the data spaces 212 of each of the multiple data providers 200. The determination step S104 is a step in which the data generation unit 20 determines whether the data spaces 212 of all the data providers 200 have been identified. If it is not determined that the data spaces 212 of all the data providers 200 have been identified, the information processing method returns to the identification step S100. If it is determined that the data spaces 212 of all the data providers 200 have been identified, the information processing method ends the identification step.

[0158] 15 is a flowchart showing an example of an information processing method according to an embodiment of the present invention. The information processing method includes an output step. The output step includes a data generation step S128 and a data output step S140. The output step may further include a target value acquisition step S108, a candidate generation step S112, a distance d calculation step S116, a target variable generation step S119, a determination step S124, a determination step S132, a user condition provision step S135, a user condition acquisition step S136, and a storage step S138.

[0159] The target value acquisition step S108 is a step in which the data generation unit 20 acquires the target value Vt of the generation target variable 226. Each of the multiple data providers 200 may input the target value Vt to the input unit 30. Each of the multiple users 260 may input the target value Vt to the input unit 30. The data generation unit 20 may acquire the target value Vt input by each of the multiple data providers 200 or each of the multiple users 260.

[0160] The candidate generation step S112 is a step in which the data generation unit 20 generates a candidate Ve for the optimized data Dt based on the first reference data 210 aggregated in the data aggregation step S84. Note that the candidate generation may involve randomly generating values. The distance d calculation step S116 is a step in which the data generation unit 20 calculates the distance d between the candidate Ve generated in the candidate generation step S112 and the data space 212 identified in the identification step S100.

[0161] The objective variable generation step S119 is a step in which the data generation unit 20 generates generated objective variables 226 that are objective variables corresponding to the candidates Ve, based on the reference data 201 that indicates the relationship between the reference explanatory variables 204 and the reference objective variables 206 that correspond to the reference explanatory variables 204, thereby generating generated data 220. The objective variable generation step S119 may be a step in which the data generation unit 20 generates generated data 220 that includes a plurality of candidates Ve and a plurality of generated objective variables 226 that correspond to the plurality of candidates Ve, respectively. The objective variable generation step S119 may be a step in which the data generation unit 20 generates generated data 220 by generating a generated objective variable 226 that is an objective variable that corresponds to any of the candidates Ve, based on the candidates Ve generated in the candidate generation step S112 and the distance d calculated in the distance d calculation step S116.

[0162] The determination step S124 is a step in which the data generation unit 20 determines whether to generate a new candidate Ve. The data generation unit 20 may determine not to generate a new candidate Ve if the difference between the predicted value Vp and the target value Vt is less than a predetermined threshold. The data generation unit 20 may determine not to generate a new candidate Ve if the predicted value Vp exceeds the target value Vt. The data generation unit 20 may determine not to generate a new candidate Ve if the number of candidates generated in the candidate generation step S112 reaches a predetermined number. Furthermore, the determination step S124 is a step in which the data generation unit 20 determines whether to generate a new candidate Ve based on the data space information Is. The determination step S124 may determine whether to generate a new candidate Ve based on the candidate Ve generated in the candidate generation step S112 and the distance d calculated in the distance d calculation step S116. The data generation unit 20 may determine not to generate a new candidate Ve if the distance d exceeds a predetermined value.

[0163] The data generating step S128 may be a step in which the data generating unit 20 selects optimized data Dt, which are explanatory variables corresponding to the target value Vt, based on the target value Vt and the data space information Is. The optimized data Dt or the generated data 220 selected in the data generating step S128 may be output by the data output unit 40 to the terminal 300 of the data provider 200 or the terminal 360 of the user 260.

[0164] The data generation step S128 may be a step in which the data generation unit 20 selects one of the candidates Ve as the optimized data Dt based on the predicted value Vp predicted in the predicted value prediction step S120 and the target value Vt acquired in the target value acquisition step S108.

[0165] The identification step S100 (see FIG. 14 ) may be a step in which the identification unit 10 identifies the first data space 212-1 of the first data provider 200-1. The candidate generation step S112 may be a step in which the data generation unit 20 generates a first candidate Ve1 for the optimized data Dt for the first data provider 200-1. The distance d calculation step S116 may be a step in which the data generation unit 20 calculates a first distance d1 between the first candidate and the first data space 212-1. The data generation step S128 may be a step in which the data generation unit 20 recommends the optimized data Dt based on the first distance d1.

[0166] The identification step S100 (see FIG. 14 ) may be a step in which the identification unit 10 identifies the second data space 212-2 of the second data provider 200-2. The distance d calculation step S116 may be a step in which the data generation unit 20 calculates a second distance d2 between the first candidate Ve1 of the optimized data Dt from the first data provider 200-1 and the second data space 212-2. The data generation step S128 may be a step in which the data generation unit 20 recommends the optimized data Dt further based on the second distance d2.

[0167] The data generating step S128 may be a step in which the data generating unit 20 determines the optimized data Dt such that the first distance d1 is equal to or greater than the first threshold distance. The data generating step S128 may be a step in which the data generating unit 20 determines the optimized data Dt such that the second distance d2 is equal to or less than the second threshold distance. The data generating step S128 may be a step in which the data generating unit 20 determines the optimized data Dt such that the first distance d1 is equal to or greater than the first threshold distance and the second distance d2 is equal to or less than the second threshold distance.

[0168] In the determination step S132, the data generating unit 20 determines whether the generation of the optimized data Dt has been completed for all target values ​​Vt. If it is determined that the generation of the optimized data Dt has been completed, the information processing method proceeds to the user condition providing step S135. If it is not determined that the generation of the optimized data Dt has been completed, the information processing method returns to the target value obtaining step S108. In the target value obtaining step S108, the data generating unit 20 may obtain target values ​​Vt for which the generation of the optimized data Dt has not been completed.

[0169] The user condition providing step S135 is a step in which the user 260 provides the user conditions 262 for acquiring the generation explanatory variables 224. The user condition providing step S135 may be a step in which the user 260 registers the user conditions 262 in the intermediation unit 420 via the terminal 360. The user condition acquiring step S136 is a step in which the data generation unit 20 acquires the user conditions 262 presented by the user 260.

[0170] The data output step S140 is a step in which the data output unit 40 generates output data 228 including the generation explanatory variables 224 based on the generation objective variables 226 and predetermined output conditions Ca, and outputs the generated output data 228. The data output step S140 may be a step in which the data output unit 40 outputs the generation objective variables 226 to multiple users 260 of the generation data 220, and outputs the generation explanatory variables 224 to at least one user 260 based on the output conditions Ca and user conditions 262 presented by each user 260 in order to obtain the generation explanatory variables 224 corresponding to the output generation objective variables 226.

[0171] The data output step S140 may be a step in which the data output unit 40 outputs the output data 228 to the intermediation unit 420 of the user 260 who has presented the user condition 262 that satisfies the output condition Ca. The data output step S140 may include a step in which the terminal 360 of the user 260 notifies the user 260 that the output data 228 has been output to the intermediation unit 420. The user 260 who has received the notification can obtain the output data 228 by accessing the intermediation unit 420 via the terminal 360.

[0172] The data output step S140 may be a step in which, when the data output unit 40 outputs a generation explanatory variable 224 to at least one user 260 whose user conditions 262 presented by the user 260 to obtain the generation explanatory variable 224 satisfy the output condition Ca, the data output unit 40 does not output the generation target variable 226 corresponding to the output generation explanatory variable 224 to other users 260 for a predetermined period of time.

[0173] The data output step S140 may be a step in which the data output unit 40 does not output the output generation explanatory variable 224 and other generation explanatory variables 224 that are within a predetermined similarity range with respect to the output generation explanatory variable 224 to other users 260 for a predetermined period of time. The data generation step S128 may be a step in which the data generation unit 20 determines the above-mentioned similarity range based on the similarity between one generation explanatory variable 224 and other generation explanatory variables 224. The data generation step S128 may be a step in which the data generation unit 20 determines the above-mentioned similarity range based on the similarity between one generation explanatory variable 224 and other generation explanatory variables 224 and the similarity between one generation target variable 226 and other generation target variables 226.

[0174] The storage step S138 is a step in which the storage unit 50 stores the user conditions 262. The data output step S140 may be a step in which the data output unit 40 outputs the generation explanatory variables 224 to at least one user 260 when the user conditions 262 of at least one user 260 satisfy the output condition Ca for the generation objective variables 226 that at least one user 260 wishes to acquire.

[0175] The user conditions 262 may include the amount that the user 260 will pay to acquire the generation objective variable 226 and the generation explanatory variable 224. The data output step S140 may be a step in which, when the user conditions 262 of each of the multiple users 260 satisfy the output condition Ca, the data output unit 40 determines one user 260 to which the generation explanatory variable 224 is to be output, based on the amount that the user 260 will pay.

[0176] The data output step S140 may be a step in which the data output unit 40 conceals the user conditions 262 of one user 260 from other users 260. The data output step S140 may be a step in which the data output unit 40 outputs, to the one user 260, information indicating other generation objective variables 226 that are recommended to be acquired, based on the generation objective variables 226 corresponding to the generation explanatory variables 224 output to the one user 260 in the generation data 220.

[0177] For one user 260 of the generated data 220, one exclusive range for the generation object variable 226 may be determined in advance. For another user 260 of the generated data 220, another exclusive range for the generation object variable 226 may be determined in advance. The data output step S140 may be a step in which, when the generation object variable 226 that one user 260 wishes to acquire belongs to another exclusive range, the data output unit 40 does not output the generation object variable 226 that belongs to the other exclusive range to the one user 260. The data output step S140 may be a step in which, when the generation object variable 226 that one user 260 wishes to acquire belongs to the other exclusive range, the data output unit 40 determines that the generation object variable 226 that belongs to the other exclusive range does not satisfy the output condition Ca.

[0178] The data output step S140 may be a step in which, when the data output unit 40 outputs the generation explanatory variable 224 to the user 260 of the generation data 220, a reward is output to the data provider 200 who provided the individual explanatory variable 203 based on the distance between the output generation explanatory variable 224 and the individual explanatory variable 203.

[0179] 16 is a flowchart illustrating another example of an information processing method according to an embodiment of the present invention. This information processing method differs from the information processing method illustrated in FIG. 15 in that it does not include a response variable generation step S119 and further includes a predicted value prediction step S120.

[0180] The predicted value prediction step S120 is a step in which the first inference model 22 predicts a predicted value Vp of the generation target variable 226 corresponding to the candidate Ve in the optimized data Dt. The determination step S124 may be a step in which the data generation unit 20 determines not to generate a new candidate Ve when the difference between the predicted value Vp and the target value Vt is less than a predetermined threshold. The determination step S124 may be a step in which the data generation unit 20 determines not to generate a new candidate Ve when the predicted value Vp exceeds the target value Vt. The determination step S124 may be a step in which the data generation unit 20 determines not to generate a new candidate Ve when the number of times candidates have been generated in the candidate generation step S112 has reached a predetermined number.

[0181] If it is determined in the determination step S124 that the prediction of the predicted value Vp is to be ended, the information processing method proceeds to the data generation step S128. In the information processing method of this example, the output step ends in the data output step S140.

[0182] 17 is a flowchart showing another example of an information processing method according to an embodiment of the present invention. The information processing method may include a first model generation step. The first model generation step may include a first inference model generation step S200, a model storage step S202, and a determination step S204.

[0183] The first inference model generation step S200 is a step in which the data generation unit 20 performs machine learning on the first reference data 210 to generate a first inference model 22 that outputs a first predicted value Vp1 of the generated objective variable 226 when an input explanatory variable is input. The model storage step S202 is a step in which the storage unit 50 stores the first inference model 22.

[0184] The determination step S204 is a step in which the data generation unit 20 determines whether machine learning has been performed for all objective variables (in the examples of Figures 3 and 12, "hardness" and "stretchability" in evaluation step E). If it is determined that machine learning has not been performed for all objective variables, the information processing method returns to the first inference model generation step S200. If it is determined that machine learning has been performed for all objective variables, the information processing method ends the first inference model generation step.

[0185] 18 is a flowchart showing another example of an information processing method according to an embodiment of the present invention. The information processing method may include a second model generation step. The second model generation step may include a data exclusion step S88, a second inference model generation step S240, a model storage step S242, and a judgment step S244.

[0186] The data exclusion step S88 is a step in which the data generation unit 20 excludes individual reference data 202 corresponding to at least one of the multiple data providers 200 from the first reference data 210. By excluding one individual reference data 202 from the first reference data 210 in the data exclusion step S88, second reference data 250 is generated.

[0187] The second inference model generation step S240 is a step in which the data generation unit 20 performs machine learning on the second reference data 250 to generate a second inference model 24 that outputs a second predicted value Vp2 of the generated objective variable 226 when an input explanatory variable is input. The model storage step S242 is a step in which the memory unit 50 stores the second inference model 24.

[0188] The determination step S244 is a step in which the data generation unit 20 determines whether machine learning has been performed for all objective variables (in the examples of Figures 3 and 12, "hardness" and "stretchability" in evaluation step E). If it is determined that machine learning has not been performed for all objective variables, the information processing method returns to the second inference model generation step S240. If it is determined that machine learning has been performed for all objective variables, the information processing method terminates the second inference model generation step.

[0189] 19 is a flowchart showing another example of an information processing method according to an embodiment of the present invention. The information processing method may include a contribution calculation step. The contribution calculation step may include a contribution calculation step S150, a determination step S152, and a reward output step S154.

[0190] The contribution calculation step S150 is a step in which the contribution calculation unit 60 calculates, based on the first predicted value Vp1 and the second predicted value Vp2, the contribution of one individual reference data 202 corresponding to at least one data provider 200 to the generation explanatory variable 224. In a case in which the reference explanatory variable 204 has multiple types of parameters (15 types in the example of FIG. 12 , from “raw material A” to “reference objective variable 206” of evaluation condition E) and at least one type of parameter is provided by at least one data provider 200, the contribution calculation step S130 may be a step in which the contribution calculation unit 60 calculates the contribution of at least one parameter to the optimization data Dt based on the optimization data Dt and the first reference data 210.

[0191] In the determination step S152, the data generation unit 20 determines whether the calculation of the contributions to the optimized data Dt has been completed for all data providers 200. If it is determined that the calculation of the contributions has been completed, the information processing method proceeds to the reward output step S154. If it is not determined that the calculation of the contributions has been completed, the information processing method returns to the contribution calculation step S150.

[0192] The reward output step S154 is a step in which, when the data output unit 40 outputs the generation explanatory variable 224 to the user 260 of the generation data 220, the data output unit 40 outputs a reward to each of the multiple data providers 200 in accordance with the degree of contribution of each of the multiple data providers 200 to the output generation explanatory variable 224.

[0193] 20 is a flowchart showing another example of an information processing method according to an embodiment of the present invention. The information processing method may include a third model generation step. The third model generation step may include a data exclusion step S88, a third inference model generation step S280, a model storage step S282, and a judgment step S284. The data exclusion step S88 is the same as in FIG. 18, so its description will be omitted.

[0194] The third inference model generation step S280 is a step in which the data generation unit 20 performs machine learning on third reference data 230 obtained by excluding one reference explanatory variable 204 and one reference objective variable 206 corresponding to the one reference explanatory variable from the first reference data 210, thereby generating a third inference model 26 that outputs a third predicted value Vp3 of the generated objective variable 226 when an input explanatory variable is input. The model storage step S282 is a step in which the storage unit 50 stores the third inference model 26.

[0195] Similar to the determination step S204, the determination step S284 is a step in which the data generation unit 20 determines whether machine learning has been performed for all dependent variables. If it is determined that machine learning has not been performed for all dependent variables, the information processing method returns to the third inference model generation step S280. If it is determined that machine learning has been performed for all dependent variables, the information processing method ends the third inference model generation step.

[0196] 21 is a flowchart showing another example of an information processing method according to an embodiment of the present invention. The information processing method may include a reliability evaluation step S160 and a generated target variable output step S162.

[0197] The reliability evaluation step S160 is a step in which the reliability evaluation unit 70 evaluates the reliability of the first reference data 210 based on the first predicted value Vp1 and the third predicted value Vp3. The reliability evaluation step S160 may be a step in which the reliability evaluation unit 70 evaluates whether the reliability of the first reference data 210 is in one of a plurality of levels based on the difference between the first predicted value Vp1 and the third predicted value Vp3 and one or more reliability thresholds.

[0198] The generated target variable output step S162 is a step in which the data output unit 40 decides whether or not to output the generated target variable 226 generated by the data generation unit 20 to the user 260 of the generated data 220 based on the reliability evaluated in the reliability evaluation step S160.

[0199] Reference data may include experimental data obtained through experiments. When formulating a plan for collecting experimental data to be used in materials informatics, one of the challenges can be the insufficiency or excess of explanatory variables required for predicting physical properties. For high-performance materials manufactured through complex, multi-step processes, even the specialized knowledge of experienced personnel leaves the mechanism of action unclear. Therefore, it is impossible to identify the factors essential for achieving the target performance. This can lead to the aforementioned challenges. In cases where data is collected across multiple processes, such as the manufacturing and preparation of materials, as well as their processing and treatment, different people may collect experimental data for each process. In such cases, it is difficult to grasp the entire data spanning multiple processes, making the aforementioned challenges particularly pronounced.

[0200] In order to prevent over- or under-collection of data under such conditions where it is difficult for each entity to grasp the entire experimental data, one embodiment of the present invention introduces an index called the degree of contribution of experimental data to each output data. The degree of contribution in this invention is an index that indicates the degree of importance of existing experimental data for recommending experimental conditions by a computer using, for example, materials informatics.

[0201] Furthermore, in one embodiment of the present invention, a plan for collecting additional experimental data is facilitated based on the contribution of existing experimental data to certain optimization data, and data is collected from an experimental device based on the plan. The information processing device of the present invention may include at least one of a "data generation unit," a "contribution calculation unit," a "planning unit," and a "control unit." The "data generation unit" may have an inference model and, when a target value of a generated objective variable is input, output optimization data, which is a generated explanatory variable corresponding to the target value. The "contribution calculation unit" may calculate the contribution of at least one experimental data item included in the reference data to the optimization data based on the optimization data and reference data (teaching data used in machine learning of the inference model). The "planning unit" may formulate and output an additional experimental data collection plan based on the contribution. The "control unit" may acquire experimental data by controlling the experimental device based on the data collection plan.

[0202] 22 is a diagram showing another example configuration of the information processing device 100. The information processing device 100 of this example includes a data generation unit 20, a contribution calculation unit 60, a plan formulation unit 72, and a control unit 74. The information processing device 100 may further include one or more, or all, of the identification unit 10, the input unit 30, the data output unit 40, the storage unit 50, and the reliability evaluation unit 70. Components in FIG. 22 that are assigned the same reference numerals as those in FIG. 4 may have the same functions and configurations as any of the examples described in FIGS. 1 to 21.

[0203] The information processing device 100 of this example facilitates planning the collection of additional experimental data based on the contribution of existing experimental data to certain optimization data, and collects data from experimental equipment based on the plan. The data generation unit 20 has a first inference model 22, and when a target value of a generation objective variable is input, outputs optimization data that is a generation explanatory variable corresponding to the target value. The operation of the data generation unit 20 may be similar to any of the examples described in Figures 1 to 21.

[0204] The contribution calculation unit 60 may calculate the contribution of at least one piece of experimental data included in the first teacher data to the optimization data based on the optimization data and the first teacher data (teacher data used in the machine learning of the first inference model). The operation of the contribution calculation unit 60 may be similar to any of the examples described with reference to FIGS. 1 to 21 .

[0205] The contribution calculation unit 60 may calculate the contribution of each type of explanatory variable included in the experimental data. The type of explanatory variable may be the type, shape, size, temperature in each process, time of each process, etc. of the input material. The type of explanatory variable for which the contribution is to be calculated is set as the target type. The contribution calculation unit 60 can calculate the contribution of the explanatory variable of the target type based on the distance between optimized data generated from reference data that excludes the explanatory variables of the target type and optimized data generated from reference data that does not exclude the explanatory variables of the target type. The specific method for calculating the contribution is the same as the examples of FIGS. 1 to 21 . The contribution calculation unit 60 may calculate the contribution of each type of explanatory variable by processing each type of explanatory variable in turn as the target type.

[0206] The plan formulation unit 72 formulates and outputs an additional experimental data collection plan based on the contribution calculated by the contribution calculation unit 60. For example, the plan formulation unit 72 formulates an experimental data collection plan for conducting an additional experiment that includes a type of explanatory variable with a high contribution. The plan formulation unit 72 may formulate an experimental data collection plan for conducting multiple additional experiments in which the values ​​of the type of explanatory variable are changed. The experimental data collection plan may specify at least one type of explanatory variable in the additional experiment to be performed, and the value or tolerance range of the explanatory variable. The experimental data collection plan may also specify all types of explanatory variables in the additional experiment to be performed, and the value or tolerance range of each explanatory variable.

[0207] The control unit 74 may acquire experimental data by controlling the experimental equipment based on the data collection plan. The control unit 74 may generate control data for controlling the experimental equipment based on the data collection plan and send it to the experimental equipment. The experimental equipment may have an execution unit that causes the experimental equipment to execute an experiment in accordance with the control data. The control unit 74 may cause the experimental equipment to execute an experiment in accordance with the data collection plan by sending the data collection plan to the experimental equipment. The experimental equipment may have an execution unit that causes the experimental equipment to execute an experiment in accordance with the data collection plan. In this case, the execution unit converts the data collection plan into control data for the experimental equipment.

[0208] The control unit 74 acquires experimental data, which is the result of an experiment performed by the experimental device based on the data collection plan. The experimental data may include explanatory variables and target variables. The control unit 74 may provide the newly acquired experimental data to the data generation unit 20. The data generation unit 20 may generate each inference model based on the reference data to which the new experimental data has been added. This type of control makes it possible to acquire additional experimental data with a high degree of contribution, and to easily generate more accurate inference models.

[0209] The contribution calculation unit 60 can specify any conditions for output based on the experimental data collection strategy. In this example, the contribution calculation unit 60 calculates the contribution of each explanatory variable to determine whether the level of a factor in the experiment is excessive or insufficient. The contribution can be output in a format that satisfies any privacy level (described below).

[0210] The following are examples of specific methods for calculating and presenting the contribution of explanatory variables. When maximizing disclosure to each user, the contribution of all explanatory variables may be presented to each user. The type of explanatory variables for which contributions are to be disclosed may be selected depending on the privacy level set. For example, when a high level of privacy is required, only values ​​related to the explanatory variables provided by each user may be presented, while other explanatory variables may be kept confidential. To utilize the relative importance of each explanatory variable, the contribution calculation unit 60 may calculate the relative ranking or deviation of the contribution of each explanatory variable. A more specific example of a calculation algorithm is as follows: (1) Determine one or more types of explanatory variables from which information related to contributions is obtained. The type of explanatory variable may be determined by one or more data providers 200. The range of information to be disclosed to each data provider 200 may be set from the following (A) to (D): (A) Disclose the actual values ​​of the contributions of all explanatory variables to all data providers 200. (B) The processed values ​​of the contributions of all explanatory variables are disclosed to all data providers 200. The processed values ​​are relative values ​​such as rankings. (C) The processed values ​​of the contributions of explanatory variables included in the experimental data provided by each data provider 200 are disclosed to that data provider 200. Furthermore, the processed values ​​of the contributions of explanatory variables whose names are concealed are disclosed. (D) The processed values ​​of the contributions of only explanatory variables included in the experimental data provided by that data provider 200 are disclosed to each data provider 200. (2): Each data provider 200 provides experimental data to be used as reference data (first teacher data). The data generation unit 20 generates a first inference model from the experimental data. (3): The type and value of the generation target variable are obtained from the user 260 who wishes to obtain optimized data. (4): Using the first inference model generated in (2), optimized data corresponding to the generation target variable in (3) is generated. (5) The contribution of each explanatory variable to the optimized data is calculated. (6): The contribution value of each explanatory variable is processed according to the type and scope of the explanatory variables to be disclosed set in (1).For example, the contribution values ​​may be processed by at least one of the following processes (E) to (G): (E) Ranking process: The contributions are sorted in descending order, and ordinal numbers are set as 1, 2, 3, etc., starting from the highest. (F) Standard deviation process: Each contribution is fitted to a normal distribution. The normal distribution may have an average of 50 and a standard deviation of 10. (G) Percentile process: The magnitude of the contribution is converted into a percentile ranging from 0 to 100 based on a preset standard. The magnitude of the contribution corresponding to percentiles 0 and 100 may be set, for example, from past actual values. For example, of the contributions calculated in the past, the maximum value may correspond to percentile 100, and the minimum value may correspond to percentile 0. (7): The names of the explanatory variables set in (1) that are not to be disclosed are anonymized. The anonymization of the names may be a process of replacing the names of the explanatory variables with blanks or preset common names, etc.

[0211] Fig. 23 shows an example of information on the contribution of each explanatory variable calculated by the contribution calculation unit 60. Fig. 23 shows information presented to one of the data providers 200. In the example of Fig. 23, the contribution of each explanatory variable is converted into an ordinal number indicating the rank. The greater the contribution of an explanatory variable, the lower the rank it is given. In addition, the names of some explanatory variables are concealed (masked data).

[0212] 24 shows another example of information on the contribution of each explanatory variable calculated by the contribution calculation unit 60. In the example of FIG. 24, the contribution of each explanatory variable is replaced with a percentile ranging from 0 to 100. For example, the greater the contribution, the closer the percentile is to "100," and the smaller the contribution, the closer it is to "0."

[0213] The planning unit 72 recommends the creation of explanatory variables to be included in the additionally acquired experimental data or the deletion of explanatory variables unnecessary for the experimental data, based on the contribution of each explanatory variable calculated by the contribution calculation unit 60. This reduces the uncertainty of predictions in the optimization data. By controlling the selection of appropriate explanatory variables in the additionally executed experiments, the planning unit 72 can acquire experimental data that can contribute to reducing the uncertainty of predictions in the first inference model used to create the optimization data.

[0214] An example of the operation of the planning unit 72 is as follows: (1): Set a reference value for the evaluation value for the accuracy of the optimization data creation of the inference model. The evaluation value will be described later in (7). (2): Set the number or distance of the target subset. (3): Set the generation objective variable. The generation objective variable may be set as a value or a range. For example, the value of the objective variable "heat resistance" is set to "150°C or higher, 200°C or lower." The generation objective variable may be input by the user 260. (4): Create a list in which each explanatory variable included in the optimization data output by the first inference model according to the generation objective variable is sorted in descending order by contribution. In other words, sort each explanatory variable in descending order by contribution. (5): Select a set number of explanatory variables from the list of explanatory variables, starting with those with the highest contribution. The value of the selection number may be set in advance by the data provider 200, etc. An inference model is created using only the selected explanatory variables in the entire first training data. (6): From the first teacher data, teacher data near the generation objective variable is extracted based on the criteria set in (2). For example, all teacher data whose distance from the objective variable to the generation objective variable is equal to or less than the distance set in (2) may be extracted. Alternatively, the number of teacher data set in (2) may be extracted in order of the distance from the objective variable to the generation objective variable, starting with the closest. (7): The inference model created in (5) is applied to each teacher data extracted in (6). For example, the explanatory variables of each teacher data are input into the inference model to calculate an objective variable for inference. An evaluation value of the inference model is calculated based on the difference between the calculated objective variable for inference and the objective variable included in the teacher data. For example, the smaller the difference, the higher the evaluation value. The evaluation value may be the difference between the value of the objective variable for inference and the value of the objective variable in the teacher data, or an evaluation value for the prediction, such as a coefficient of determination. The coefficient of determination may be, for example, the square of the correlation coefficient between the value of the objective variable for inference and the value of the objective variable in the teacher data. (8): The evaluation value calculated in (7) is compared with the reference value set in (1). If the evaluation value of the inference model is lower than the reference value, the number of selections in (5) is increased by one to create an inference model and calculate the evaluation value.The number of selections is increased by one each time, and the creation and evaluation of the inference model is repeated until the evaluation value becomes equal to or greater than the reference value. When the evaluation value becomes equal to or greater than the reference value, explanatory variables not selected in (5) are subject to an over-provision warning. If the evaluation value of the inference model created when all explanatory variables are selected in (5) is smaller than the reference value, a warning of insufficient explanatory variables may be presented. If the evaluation value of the inference model is higher than the reference value, the number of selections in (5) is reduced by one, an inference model is created, and the evaluation value is calculated. The number of selections is reduced by one each time, and the creation and evaluation of the inference model are repeated until the evaluation value becomes smaller than the reference value. Explanatory variables not selected in (5) in the processing cycle immediately preceding the processing cycle in which the evaluation value became smaller than the reference value may be subject to an over-provision warning.

[0215] The plan formulation unit 72 may generate a data collection plan to acquire additional experimental data excluding explanatory variables that are the subject of an overprovision warning. The plan formulation unit 72 may generate an experimental plan for the experimental device in which fixed values ​​are set for explanatory variables that are the subject of an overprovision warning. When a warning for insufficient explanatory variables has occurred, the plan formulation unit 72 may generate an experimental plan for the experimental device that includes new explanatory variables.

[0216] 25 is a diagram illustrating an example of the operation of the plan development unit 72. As shown in FIG. 23, the plan development unit 72 in this example determines whether or not each explanatory variable ranked by contribution is subject to an over-provision warning. In the example of FIG. 25, explanatory variables ranked 84th and above are subject to an over-provision warning. The plan development unit 72 may create a data collection plan so that additional experimental data showing diverse values ​​can be acquired for explanatory variables that are not subject to an over-provision warning.

[0217] FIG. 26 is a diagram illustrating another example of the operation of the planning unit 72. In this example, the contribution level is converted into a percentile, as shown in FIG. 24 . In this example, the evaluation value of the inference model using all explanatory variables does not reach the reference value. In this case, the planning unit 72 may issue an insufficient provision warning for all explanatory variables. The planning unit 72 may create a data collection plan so that additional experimental data can be acquired. The planning unit 72 may create a warning as part of the data collection plan, indicating that experimental data is insufficient and should be added. The planning unit 72 may issue a warning indicating that experimental data including a new explanatory variable should be added, or may specify an explanatory variable with a high contribution level for existing explanatory variables and issue a warning indicating that experimental data including the explanatory variable should be added. An explanatory variable with a high contribution level may be, for example, an explanatory variable whose contribution level or ranking, or other indicator, exceeds a set threshold.

[0218] Fig. 27 is a diagram showing another example configuration of the information processing device 100. The information processing device 100 of this example has a data acquisition unit 76 instead of the control unit 74 in the configuration described in Fig. 22. The configuration other than the data acquisition unit 76 may be the same as the example described in Fig. 22.

[0219] The planning unit 72 in this example outputs the formulated experimental data collection plan to the experimental equipment and causes the experimental equipment to execute an experiment in accordance with the experimental data collection plan. The experimental data collection plan may be similar to the example described in FIG. 22 . The planning unit 72 may output control data for each device of the experimental equipment to the experimental equipment in order to conduct an experiment in accordance with the experimental data collection plan. In response to a request from a terminal of any data provider 200, the planning unit 72 may transmit the experimental data collection plan or the control data to the terminal. For example, the data provider 200 may request the experimental data collection plan or the control data from the information processing device 100 by selecting a data request button displayed on the terminal.

[0220] The data acquisition unit 76 acquires additional experimental data from the experimental equipment that has operated based on the data collection plan or control data. The additional experimental data is similar to the example described in Figure 22. This configuration also makes it possible to acquire additional experimental data with a high degree of contribution, making it easy to generate a more accurate inference model.

[0221] 28 is a diagram showing an example of a computer 2200 in which the information processing device 100 according to an embodiment of the present invention may be embodied, in whole or in part. A program installed on the computer 2200 can cause the computer 2200 to perform operations associated with the information processing device 100 according to an embodiment of the present invention, or to function as one or more sections of the information processing device 100, or to execute the operations or one or more sections, or to execute each step of the information processing method of the present invention (see FIGS. 1 to 27). The program can be executed by the CPU 2212 to cause the computer 2200 to perform specific operations associated with some or all of the blocks in the flowcharts ( FIGS. 13 to 21 ) and block diagrams ( FIGS. 4 , 22 , and 27 ) described herein.

[0222] A computer 2200 according to one embodiment of the present invention includes a CPU 2212, a RAM 2214, a graphics controller 2216, and a display device 2218. The CPU 2212, the RAM 2214, the graphics controller 2216, and the display device 2218 are connected to one another by a host controller 2210. The computer 2200 further includes input / output units such as a communication interface 2222, a hard disk drive 2224, a DVD-ROM drive 2226, and an IC card drive. The communication interface 2222, the hard disk drive 2224, the DVD-ROM drive 2226, and the IC card drive are connected to the host controller 2210 via an input / output controller 2220. The computer further includes legacy input / output units such as a ROM 2230 and a keyboard 2242. The ROM 2230, the keyboard 2242, and the like are connected to the input / output controller 2220 via an input / output chip 2240.

[0223] The CPU 2212 controls each unit by operating in accordance with programs stored in the ROM 2230 and the RAM 2214. The graphics controller 2216 acquires image data generated by the CPU 2212 into a frame buffer or the like provided in the RAM 2214, or into the RAM 2214, so that the image data can be displayed on the display device 2218.

[0224] The communication interface 2222 communicates with other electronic devices via a network. The hard disk drive 2224 stores programs and data used by the CPU 2212 in the computer 2200. The DVD-ROM drive 2226 reads programs or data from the DVD-ROM 2201 and provides the read programs or data to the hard disk drive 2224 via the RAM 2214. The IC card drive reads programs and data from an IC card or writes programs and data to an IC card.

[0225] The ROM 2230 stores a boot program or the like that is executed by the computer 2200 upon activation, or a program that depends on the hardware of the computer 2200. The input / output chip 2240 may connect various input / output units to the input / output controller 2220 via a parallel port, a serial port, a keyboard port, a mouse port, or the like.

[0226] The programs are provided by a computer-readable medium such as a DVD-ROM 2201 or an IC card. The programs are read from the computer-readable medium, installed in the hard disk drive 2224, RAM 2214, or ROM 2230, which are also examples of computer-readable media, and executed by the CPU 2212. Information processing described in these programs is read by the computer 2200, and brings about cooperation between the programs and the various types of hardware resources described above. An apparatus or method may be configured by implementing information manipulation or processing in accordance with the use of the computer 2200.

[0227] For example, when communication is performed between computer 2200 and an external device, CPU 2212 executes a communication program loaded into RAM 2214 and commands communication processing to communication interface 2222 based on the processing described in the communication program. Under the control of CPU 2212, communication interface 2222 reads transmission data stored in a transmission buffer processing area provided in RAM 2214, hard disk drive 2224, DVD-ROM 2201, or a recording medium such as an IC card, and transmits the read transmission data to the network, or writes received data received from the network to a reception buffer processing area or the like provided on the recording medium.

[0228] The CPU 2212 may read all or a necessary portion of a file or database stored on an external recording medium such as the hard disk drive 2224, the DVD-ROM drive 2226 (DVD-ROM 2201), an IC card, etc. into the RAM 2214. The CPU 2212 may perform various types of processing on the data on the RAM 2214. The CPU 2212 may then write the processed data back to the external recording medium.

[0229] Various types of information, such as various types of programs, data, tables, and databases, may be stored on the recording medium and processed. The CPU 2212 may perform various types of processing on data read from the RAM 2214, including various types of operations, information processing, conditional decisions, conditional branches, unconditional branches, information search or replacement, etc., specified by the instruction sequences of the programs described in this disclosure. The CPU 2212 may write the results back to the RAM 2214.

[0230] The CPU 2212 may search for information in a file, database, etc. in the recording medium. For example, when a plurality of entries each having an attribute value of a first attribute associated with an attribute value of a second attribute are stored in the recording medium, the CPU 2212 searches the plurality of entries for an entry that matches a condition specified by the attribute value of the first attribute, reads the attribute value of the second attribute stored in the entry, and by reading the second attribute value, obtains the attribute value of the second attribute associated with the first attribute that satisfies a predetermined condition.

[0231] The above-described programs or software modules may be stored on the computer 2200 or in a computer-readable medium of the computer 2200. A recording medium such as a hard disk or RAM provided in a server system connected to a dedicated communication network or the Internet can be used as the computer-readable medium. The programs may be provided to the computer 2200 by the recording medium.

[0232] Although the present invention has been described above using embodiments, the technical scope of the present invention is not limited to the scope described in the above embodiments. It will be apparent to those skilled in the art that various modifications and improvements can be made to the above embodiments. It is clear from the claims that such modifications and improvements can also be included within the technical scope of the present invention.

[0233] It should be noted that the order of execution of each process, such as operations, procedures, steps, and stages, in the devices, systems, programs, and methods shown in the claims, specifications, and drawings is not specifically stated as "before," "prior to," etc., and that the processes can be performed in any order unless the output of a previous process is used in a subsequent process. Even if the operational flow in the claims, specifications, and drawings is described using "first," "next," etc. for convenience, this does not mean that the processes must be performed in this order.

[0234] 10... Identification unit, 20... Data generation unit, 21... Estimation model, 22... First inference model, 24... Second inference model, 26... Third inference model, 30... Input unit, 40... Data output unit, 50... Storage unit, 60... Contribution calculation unit, 70... Reliability evaluation unit, 100... Information processing device, 200... Data provider, 201... Reference data, 202... Individual reference data, 203... Individual explanatory variable, 204... Reference explanatory variable, 205... Individual objective variable, 206... Reference objective variable, 208... Individual reference data, 210... Reference data, 212... Data space, 214... Data space, 220... Generated data, 224... Generated explanatory variable, 226... Generated objective variable, 228... Output data, 230...reference data, 250...reference data, 260...user, 262...user conditions, 300...terminal, 360...terminal, 400...network, 420...intermediary unit, 500...information provision system, 2200...computer, 2201...DVD-ROM, 2210...host controller, 2212...CPU, 2214...RAM, 2216...graphics controller, 2218...display device, 2220...input / output controller, 2222...communication interface, 2224...hard disk drive, 2226...DVD-ROM drive, 2230...ROM, 2240...input / output chip, 2242...keyboard

Claims

1. An information processing device comprising: a data generation unit that generates generation data including generation explanatory variables and generation objective variables based on reference data that indicates the relationship between reference explanatory variables and reference objective variables that correspond to the reference explanatory variables; and a data output unit that generates output data including the generation explanatory variables based on the generation objective variables and predetermined output conditions, and outputs the generated output data.

2. The information processing device according to claim 1, wherein the data output unit outputs the generation objective variables to a plurality of users of the generation data, and outputs the generation explanatory variables to at least one of the users based on the user conditions presented by each of the users in order to obtain the generation explanatory variables corresponding to the output generation objective variables and the output conditions.

3. The information processing device according to claim 2, wherein when the data output unit outputs the generation explanatory variables to at least one of the users whose user conditions presented by the user to obtain the generation explanatory variables satisfy the output conditions, the data output unit does not output the generation target variables corresponding to the output generation explanatory variables to other of the users for a predetermined period of time.

4. The information processing device according to claim 3, wherein the data generation unit generates the generation data including a plurality of the generation explanatory variables and a plurality of the generation target variables corresponding to each of the plurality of the generation explanatory variables, and the data output unit does not output the outputted generation explanatory variables and other generation explanatory variables within a predetermined similarity range to the outputted generation explanatory variables to other users during the period.

5. The information processing device according to claim 4, wherein the data generation section determines the similarity range based on the degree of similarity between one of the generation explanatory variables and another of the generation explanatory variables.

6. The information processing device according to claim 5, wherein the data generation section determines the similarity range further based on the degree of similarity between one of the generation target variables and another of the generation target variables.

7. An information processing device as described in claim 1, further comprising a memory unit that stores user conditions presented by each of multiple users of the generation data in order to obtain the generation objective variables, and wherein the data output unit outputs the generation explanatory variables to at least one of the users when the user conditions of at least one of the users satisfy the output conditions for the generation objective variables that at least one of the users wishes to obtain.

8. The information processing device according to claim 7, wherein the user conditions include an amount to be paid by the user to acquire the generation target variable and the generation explanatory variable, and when the user conditions of each of a plurality of users satisfy the output condition, the data output unit determines one of the users to output the generation explanatory variable based on the amount.

9. The information processing device according to claim 7, wherein the data output unit keeps the user conditions of one of the users secret from the other users.

10. The information processing device according to claim 1, wherein the data output unit outputs, to a user of the generated data, information indicating other generation objective variables that are recommended for acquisition based on the generation objective variables corresponding to the generation explanatory variables output to the user.

11. The information processing device of claim 1, wherein one exclusive range for the generation target variable is predetermined for one user of the generated data, another exclusive range for the generation target variable is predetermined for another user of the generated data, and if the generation target variable that the one user wishes to acquire belongs to the other exclusive range, the data output unit does not output the generation target variable that belongs to the other exclusive range to the one user.

12. The information processing device of claim 1, wherein one exclusive range for the generation target variable is predetermined for one user of the generation data, another exclusive range for the generation target variable is predetermined for another user of the generation data, and when the generation target variable that the one user wishes to acquire belongs to the other exclusive range, the data output unit determines that the generation target variable that belongs to the other exclusive range does not satisfy the output condition.

13. The information processing device according to claim 1, wherein the reference explanatory variables and the reference target variable are provided by a data provider, and when the data output unit outputs the generation explanatory variables to a user of the generation data, it outputs a reward to the data provider who provided the reference explanatory variables based on the distance between the output generation explanatory variables and the reference explanatory variables.

14. An information processing device according to any one of claims 1 to 13, wherein the data generation unit has a first inference model that, when an input explanatory variable is input, outputs a first predicted value of the generated objective variable corresponding to the input explanatory variable by machine learning the first reference data having the reference explanatory variable and the reference objective variable provided by each of a plurality of data providers, and when a target value of the generated objective variable is input to the data generation unit, the data generation unit outputs optimized data that is the generated explanatory variable corresponding to the target value.

15. The information processing device described in claim 14, wherein the data generation unit generates candidates for the optimization data in response to the input of the target value, the first inference model outputs the first predicted value corresponding to the candidate generated by the data generation unit, and the data generation unit outputs one of the candidates as the optimization data based on the first predicted value and the target value.

16. The information processing device according to claim 14, wherein the data generation unit has a second inference model that outputs a second predicted value of the generation objective variable when the input explanatory variable is input, by machine learning second reference data obtained by excluding, from the first reference data, individual explanatory variables that are the reference explanatory variables corresponding to at least one of the plurality of data providers and individual objective variables that are the reference objective variables corresponding to the at least one of the plurality of data providers, and further comprises a contribution calculation unit that calculates a contribution of the first reference data corresponding to at least one of the data providers to the generation explanatory variables based on the first predicted value and the second predicted value.

17. The information processing device according to claim 14, wherein the reference explanatory variables have a plurality of types of parameters, at least one type of the parameters is provided by at least one of the data providers, and the information processing device further comprises a contribution calculation unit that calculates the contribution of the at least one of the parameters to the optimization data based on the optimization data and first of the reference data.

18. The information processing device described in claim 16, wherein the data generation unit performs machine learning on the second reference data for each of the plurality of data providers, from which the individual explanatory variables and the individual objective variables corresponding to one of the plurality of data providers have been excluded; the second inference model outputs the second predicted value for each of the plurality of data providers; the contribution calculation unit calculates the contribution for each of the plurality of data providers based on the first predicted value and the second predicted value for each of the plurality of data providers; and the data output unit, when outputting the generation explanatory variables to a user of the generation data, outputs a reward to each of the plurality of data providers in accordance with the contribution of each of the plurality of data providers to the output generation explanatory variables.

19. The information processing device of claim 14, wherein the data generation unit has a third inference model that outputs a third predicted value of the generated target variable when the input explanatory variable is input, by machine learning third reference data obtained by excluding one of the reference explanatory variables and one of the reference target variables corresponding to one of the reference explanatory variables from the first reference data, and further comprises a reliability evaluation unit that evaluates the reliability of the first reference data based on the first predicted value and the third predicted value.

20. The information processing device according to claim 19, wherein the data output unit determines whether or not to output the generation target variable generated by the data generation unit to a user of the generated data based on the reliability.

21. An information processing device as described in any one of claims 1 to 13, further comprising an input unit, wherein when a user of the generated data inputs to the input unit a request that the data output unit output new generated explanatory variables, the data generation unit generates new generated data.

22. An information processing device according to any one of claims 1 to 13, further comprising an input unit, wherein the reference explanatory variables and the reference objective variable are provided by a plurality of data providers, and when an individual explanatory variable that is the reference explanatory variable corresponding to at least one of the plurality of data providers and an individual objective variable that is the reference objective variable corresponding to said at least one are newly input to the input unit, the data generation unit generates new generated data.

23. The information processing device according to claim 17, wherein the reference data includes at least one piece of experimental data, and further comprising a planning unit that formulates and outputs a data collection plan for collecting additional experimental data based on the contribution calculated by the contribution calculation unit.

24. The information processing device according to claim 23, further comprising a control unit that acquires the additional experimental data by controlling an experimental device based on the data collection plan.

25. The information processing device according to claim 23, further comprising a data acquisition unit that acquires the additional experimental data from experimental equipment that has operated based on the data collection plan.

26. An information processing method comprising: a data generation step in which a data generation unit generates generation data including generation explanatory variables and generation objective variables based on reference data indicating the relationship between reference explanatory variables and reference objective variables corresponding to the reference explanatory variables; and a data output step in which a data output unit generates output data including the generation explanatory variables based on the generation objective variables and predetermined output conditions, and outputs the generated output data.

27. An information processing program for causing a computer to function as an information processing device according to any one of claims 1 to 13 and 23 to 25.

Images